As you probably know there is a trend in enterprise computing towards
networked storage. This is illustrated by the emergence during the
past few years of standards like SRP (SCSI RDMA Protocol), iSCSI
(Internet SCSI) and iSER (iSCSI Extensions for RDMA). Two different
pieces of software are necessary to make networked storage possible:
initiator software and target software. As far as I know there exist
three different SCSI target implementations for Linux:
- The iSCSI Enterprise Target Daemon (IETD,
http://iscsitarget.sourceforge.net/);
- The Linux SCSI Target Framework (STGT, http://stgt.berlios.de/);
- The Generic SCSI Target Middle Level for Linux project (SCST,
http://scst.sourceforge.net/).
Since I was wondering which SCSI target software would be best suited
for an InfiniBand network, I started evaluating the STGT and SCST SCSI
target implementations. Apparently the performance difference between
STGT and SCST is small on 100 Mbit/s and 1 Gbit/s Ethernet networks,
but the SCST target software outperforms the STGT software on an
InfiniBand network. See also the following thread for the details:
http://sourceforge.net/mailarchive/forum.php?thread_name=e2e108260801170127w2937b2afg9bef324efa945e43%40mail.gmail.com&forum_name=scst-devel.
About the design of the SCST software: while one of the goals of the
STGT project was to keep the in-kernel code minimal, the SCST project
implements the whole SCSI target in kernel space. SCST is implemented
as a set of new kernel modules, only minimal changes to the existing
kernel are necessary before the SCST kernel modules can be used. This
is the same approach that will be followed in the very near future in
the OpenSolaris kernel (see also
http://opensolaris.org/os/project/comstar/). More information about
the design of SCST can be found here:
http://scst.sourceforge.net/doc/scst_pg.html.
My impression is that both the STGT and SCST projects are well
designed, well maintained and have a considerable user base. According
to the SCST maintainer (Vladislav Bolkhovitin), SCST is superior to
STGT with respect to features, performance, maturity, stability, and
number of existing target drivers. Unfortunately the SCST kernel code
lives outside the kernel tree, which makes SCST harder to use than
STGT.
As an SCST user, I would like to see the SCST kernel code integrated
in the mainstream kernel because of its excellent performance on an
InfiniBand network. Since the SCST project comprises about 14 KLOC,
reviewing the SCST code will take considerable time. Who will do this
reviewing work ? And with regard to the comments made by the
reviewers: Vladislav, do you have the time to carry out the
modifications requested by the reviewers ? I expect a.o. that
reviewers will ask to move SCST's configuration pseudofiles from
procfs to sysfs.
Bart Van Assche.
Bart Van Assche wrote:
> As you probably know there is a trend in enterprise computing towards
> networked storage. This is illustrated by the emergence during the
> past few years of standards like SRP (SCSI RDMA Protocol), iSCSI
> (Internet SCSI) and iSER (iSCSI Extensions for RDMA). Two different
> pieces of software are necessary to make networked storage possible:
> initiator software and target software. As far as I know there exist
> three different SCSI target implementations for Linux:
> - The iSCSI Enterprise Target Daemon (IETD,
> http://iscsitarget.sourceforge.net/);
> - The Linux SCSI Target Framework (STGT, http://stgt.berlios.de/);
> - The Generic SCSI Target Middle Level for Linux project (SCST,
> http://scst.sourceforge.net/).
> Since I was wondering which SCSI target software would be best suited
> for an InfiniBand network, I started evaluating the STGT and SCST SCSI
> target implementations. Apparently the performance difference between
> STGT and SCST is small on 100 Mbit/s and 1 Gbit/s Ethernet networks,
> but the SCST target software outperforms the STGT software on an
> InfiniBand network. See also the following thread for the details:
> http://sourceforge.net/mailarchive/forum.php?thread_name=e2e108260801170127w2937b2afg9bef324efa945e43%40mail.gmail.com&forum_name=scst-devel.
>
> About the design of the SCST software: while one of the goals of the
> STGT project was to keep the in-kernel code minimal, the SCST project
> implements the whole SCSI target in kernel space. SCST is implemented
> as a set of new kernel modules, only minimal changes to the existing
> kernel are necessary before the SCST kernel modules can be used. This
> is the same approach that will be followed in the very near future in
> the OpenSolaris kernel (see also
> http://opensolaris.org/os/project/comstar/). More information about
> the design of SCST can be found here:
> http://scst.sourceforge.net/doc/scst_pg.html.
>
> My impression is that both the STGT and SCST projects are well
> designed, well maintained and have a considerable user base. According
> to the SCST maintainer (Vladislav Bolkhovitin), SCST is superior to
> STGT with respect to features, performance, maturity, stability, and
> number of existing target drivers. Unfortunately the SCST kernel code
> lives outside the kernel tree, which makes SCST harder to use than
> STGT.
>
> As an SCST user, I would like to see the SCST kernel code integrated
> in the mainstream kernel because of its excellent performance on an
> InfiniBand network. Since the SCST project comprises about 14 KLOC,
> reviewing the SCST code will take considerable time. Who will do this
> reviewing work ? And with regard to the comments made by the
> reviewers: Vladislav, do you have the time to carry out the
> modifications requested by the reviewers ? I expect a.o. that
> reviewers will ask to move SCST's configuration pseudofiles from
> procfs to sysfs.
Sure, I do, although I personally don't see much sense in such move.
> Bart Van Assche.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
On Wed, 2008-01-23 at 15:22 +0100, Bart Van Assche wrote:
> As you probably know there is a trend in enterprise computing towards
> networked storage. This is illustrated by the emergence during the
> past few years of standards like SRP (SCSI RDMA Protocol), iSCSI
> (Internet SCSI) and iSER (iSCSI Extensions for RDMA). Two different
> pieces of software are necessary to make networked storage possible:
> initiator software and target software. As far as I know there exist
> three different SCSI target implementations for Linux:
> - The iSCSI Enterprise Target Daemon (IETD,
> http://iscsitarget.sourceforge.net/);
> - The Linux SCSI Target Framework (STGT, http://stgt.berlios.de/);
> - The Generic SCSI Target Middle Level for Linux project (SCST,
> http://scst.sourceforge.net/).
> Since I was wondering which SCSI target software would be best suited
> for an InfiniBand network, I started evaluating the STGT and SCST SCSI
> target implementations. Apparently the performance difference between
> STGT and SCST is small on 100 Mbit/s and 1 Gbit/s Ethernet networks,
> but the SCST target software outperforms the STGT software on an
> InfiniBand network. See also the following thread for the details:
> http://sourceforge.net/mailarchive/forum.php?thread_name=e2e108260801170127w2937b2afg9bef324efa945e43%40mail.gmail.com&forum_name=scst-devel.
That doesn't seem to pull up a thread. However, I assume it's these
figures:
.............................................................................................
. . STGT read SCST read . STGT read SCST read .
. . performance performance . performance performance .
. . (0.5K, MB/s) (0.5K, MB/s) . (1 MB, MB/s) (1 MB, MB/s) .
.............................................................................................
. Ethernet (1 Gb/s network) . 77 78 . 77 89 .
. IPoIB (8 Gb/s network) . 163 185 . 201 239 .
. iSER (8 Gb/s network) . 250 N/A . 360 N/A .
. SRP (8 Gb/s network) . N/A 421 . N/A 683 .
.............................................................................................
On the comparable figures, which only seem to be IPoIB they're showing a
13-18% variance, aren't they? Which isn't an incredible difference.
> About the design of the SCST software: while one of the goals of the
> STGT project was to keep the in-kernel code minimal, the SCST project
> implements the whole SCSI target in kernel space. SCST is implemented
> as a set of new kernel modules, only minimal changes to the existing
> kernel are necessary before the SCST kernel modules can be used. This
> is the same approach that will be followed in the very near future in
> the OpenSolaris kernel (see also
> http://opensolaris.org/os/project/comstar/). More information about
> the design of SCST can be found here:
> http://scst.sourceforge.net/doc/scst_pg.html.
>
> My impression is that both the STGT and SCST projects are well
> designed, well maintained and have a considerable user base. According
> to the SCST maintainer (Vladislav Bolkhovitin), SCST is superior to
> STGT with respect to features, performance, maturity, stability, and
> number of existing target drivers. Unfortunately the SCST kernel code
> lives outside the kernel tree, which makes SCST harder to use than
> STGT.
>
> As an SCST user, I would like to see the SCST kernel code integrated
> in the mainstream kernel because of its excellent performance on an
> InfiniBand network. Since the SCST project comprises about 14 KLOC,
> reviewing the SCST code will take considerable time. Who will do this
> reviewing work ? And with regard to the comments made by the
> reviewers: Vladislav, do you have the time to carry out the
> modifications requested by the reviewers ? I expect a.o. that
> reviewers will ask to move SCST's configuration pseudofiles from
> procfs to sysfs.
The two target architectures perform essentially identical functions, so
there's only really room for one in the kernel. Right at the moment,
it's STGT. Problems in STGT come from the user<->kernel boundary which
can be mitigated in a variety of ways. The fact that the figures are
pretty much comparable on non IB networks shows this.
I really need a whole lot more evidence than at worst a 20% performance
difference on IB to pull one implementation out and replace it with
another. Particularly as there's no real evidence that STGT can't be
tweaked to recover the 20% even on IB.
James
> . . STGT read SCST read . STGT read SCST read .
> . . performance performance . performance performance .
> . . (0.5K, MB/s) (0.5K, MB/s) . (1 MB, MB/s) (1 MB, MB/s) .
> . iSER (8 Gb/s network) . 250 N/A . 360 N/A .
> . SRP (8 Gb/s network) . N/A 421 . N/A 683 .
> On the comparable figures, which only seem to be IPoIB they're showing a
> 13-18% variance, aren't they? Which isn't an incredible difference.
Maybe I'm all wet, but I think iSER vs. SRP should be roughly
comparable. The exact formatting of various messages etc. is
different but the data path using RDMA is pretty much identical. So
the big difference between STGT iSER and SCST SRP hints at some big
difference in the efficiency of the two implementations.
- R.
On Tue, 29 Jan 2008 13:31:52 -0800
Roland Dreier <[email protected]> wrote:
> > . . STGT read SCST read . STGT read SCST read .
> > . . performance performance . performance performance .
> > . . (0.5K, MB/s) (0.5K, MB/s) . (1 MB, MB/s) (1 MB, MB/s) .
> > . iSER (8 Gb/s network) . 250 N/A . 360 N/A .
> > . SRP (8 Gb/s network) . N/A 421 . N/A 683 .
>
> > On the comparable figures, which only seem to be IPoIB they're showing a
> > 13-18% variance, aren't they? Which isn't an incredible difference.
>
> Maybe I'm all wet, but I think iSER vs. SRP should be roughly
> comparable. The exact formatting of various messages etc. is
> different but the data path using RDMA is pretty much identical. So
> the big difference between STGT iSER and SCST SRP hints at some big
> difference in the efficiency of the two implementations.
iSER has parameters to limit the maximum size of RDMA (it needs to
repeat RDMA with a poor configuration)?
Anyway, here's the results from Robin Humble:
iSER to 7G ramfs, x86_64, centos4.6, 2.6.22 kernels, git tgtd,
initiator end booted with mem=512M, target with 8G ram
direct i/o dd
write/read 800/751 MB/s
dd if=/dev/zero of=/dev/sdc bs=1M count=5000 oflag=direct
dd of=/dev/null if=/dev/sdc bs=1M count=5000 iflag=direct
http://www.mail-archive.com/[email protected]/msg13502.html
I think that STGT is pretty fast with the fast backing storage.
I don't think that there is the notable perfornace difference between
kernel-space and user-space SRP (or ISER) implementations about moving
data between hosts. IB is expected to enable user-space applications
to move data between hosts quickly (if not, what can IB provide us?).
I think that the question is how fast user-space applications can do
I/Os ccompared with I/Os in kernel space. STGT is eager for the advent
of good asynchronous I/O and event notification interfances.
One more possible optimization for STGT is zero-copy data
transfer. STGT uses pre-registered buffers and move data between page
cache and thsse buffers, and then does RDMA transfer. If we implement
own caching mechanism to use pre-registered buffers directly with (AIO
and O_DIRECT), then STGT can move data without data copies.
FUJITA Tomonori wrote:
> On Tue, 29 Jan 2008 13:31:52 -0800
> Roland Dreier <[email protected]> wrote:
>
>> > . . STGT read SCST read . STGT read SCST read .
>> > . . performance performance . performance performance .
>> > . . (0.5K, MB/s) (0.5K, MB/s) . (1 MB, MB/s) (1 MB, MB/s) .
>> > . iSER (8 Gb/s network) . 250 N/A . 360 N/A .
>> > . SRP (8 Gb/s network) . N/A 421 . N/A 683 .
>>
>> > On the comparable figures, which only seem to be IPoIB they're showing a
>> > 13-18% variance, aren't they? Which isn't an incredible difference.
>>
>> Maybe I'm all wet, but I think iSER vs. SRP should be roughly
>> comparable. The exact formatting of various messages etc. is
>> different but the data path using RDMA is pretty much identical. So
>> the big difference between STGT iSER and SCST SRP hints at some big
>> difference in the efficiency of the two implementations.
>
> iSER has parameters to limit the maximum size of RDMA (it needs to
> repeat RDMA with a poor configuration)?
>
>
> Anyway, here's the results from Robin Humble:
>
> iSER to 7G ramfs, x86_64, centos4.6, 2.6.22 kernels, git tgtd,
> initiator end booted with mem=512M, target with 8G ram
>
> direct i/o dd
> write/read 800/751 MB/s
> dd if=/dev/zero of=/dev/sdc bs=1M count=5000 oflag=direct
> dd of=/dev/null if=/dev/sdc bs=1M count=5000 iflag=direct
>
Both Robin (iser/stgt) and Bart (scst/srp) using ramfs
Robin's numbers come from DDR IB HCAs
Bart's numbers come from SDR IB HCAs:
Results with /dev/ram0 configured as backing store on the
target (buffered I/O):
Read Write Read
Write
performance performance
performance performance
(0.5K, MB/s) (0.5K, MB/s) (1 MB,
MB/s) (1 MB, MB/s)
STGT + iSER 250 48 349
781
SCST + SRP 411 66 659
746
Results with /dev/ram0 configured as backing store on the
target (direct I/O):
Read Write Read
Write
performance performance
performance performance
(0.5K, MB/s) (0.5K, MB/s) (1 MB,
MB/s) (1 MB, MB/s)
STGT + iSER 7.9 9.8 589
647
SCST + SRP 12.3 9.7 811
794
http://www.mail-archive.com/[email protected]/msg13514.html
Here are my numbers with DDR IB HCAs, SCST/SRP 5G /dev/ram0
block_io mode, RHEL5 2.6.18-8.el5
direct i/o dd
write/read 1100/895 MB/s
dd if=/dev/zero of=/dev/sdc bs=1M count=5000 oflag=direct
dd of=/dev/null if=/dev/sdc bs=1M count=5000 iflag=direct
buffered i/o dd
write/read 950/770 MB/s
dd if=/dev/zero of=/dev/sdc bs=1M count=5000
dd of=/dev/null if=/dev/sdc bs=1M count=5000
So when using DDR IB hcas:
stgt/iser scst/srp
direct I/O 800/751 1100/895
buffered I/O 1109/350 950/770
-vu
> http://www.mail-archive.com/[email protected]/msg13502.html
>
> I think that STGT is pretty fast with the fast backing storage.
>
>
> I don't think that there is the notable perfornace difference between
> kernel-space and user-space SRP (or ISER) implementations about moving
> data between hosts. IB is expected to enable user-space applications
> to move data between hosts quickly (if not, what can IB provide us?).
>
> I think that the question is how fast user-space applications can do
> I/Os ccompared with I/Os in kernel space. STGT is eager for the advent
> of good asynchronous I/O and event notification interfances.
>
>
> One more possible optimization for STGT is zero-copy data
> transfer. STGT uses pre-registered buffers and move data between page
> cache and thsse buffers, and then does RDMA transfer. If we implement
> own caching mechanism to use pre-registered buffers directly with (AIO
> and O_DIRECT), then STGT can move data without data copies.
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Scst-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scst-devel
>
On Jan 29, 2008 9:42 PM, James Bottomley
<[email protected]> wrote:
> > As an SCST user, I would like to see the SCST kernel code integrated
> > in the mainstream kernel because of its excellent performance on an
> > InfiniBand network. Since the SCST project comprises about 14 KLOC,
> > reviewing the SCST code will take considerable time. Who will do this
> > reviewing work ? And with regard to the comments made by the
> > reviewers: Vladislav, do you have the time to carry out the
> > modifications requested by the reviewers ? I expect a.o. that
> > reviewers will ask to move SCST's configuration pseudofiles from
> > procfs to sysfs.
>
> The two target architectures perform essentially identical functions, so
> there's only really room for one in the kernel. Right at the moment,
> it's STGT. Problems in STGT come from the user<->kernel boundary which
> can be mitigated in a variety of ways. The fact that the figures are
> pretty much comparable on non IB networks shows this.
Are you saying that users who need an efficient iSCSI implementation
should switch to OpenSolaris ? The OpenSolaris COMSTAR project involves
the migration of the existing OpenSolaris iSCSI target daemon from
userspace to their kernel. The OpenSolaris developers are
spending time on this because they expect a significant performance
improvement.
> I really need a whole lot more evidence than at worst a 20% performance
> difference on IB to pull one implementation out and replace it with
> another. Particularly as there's no real evidence that STGT can't be
> tweaked to recover the 20% even on IB.
My measurements on a 1 GB/s InfiniBand network have shown that the current
SCST implementation is able to read data via direct I/O at a rate of 811 GB/s
(via SRP) and that the current STGT implementation is able to transfer data at a
rate of 589 MB/s (via iSER). That's a performance difference of 38%.
And even more important, the I/O latency of SCST is significantly
lower than that
of STGT. This is very important for database workloads -- the I/O pattern caused
by database software is close to random I/O, and database software needs low
latency I/O in order to run efficiently.
In the thread with the title "Performance of SCST versus STGT" on the
SCST-devel /
STGT-devel mailing lists not only the raw performance numbers were discussed but
also which further performance improvements are possible. It became clear that
the SCST performance can be improved further by implementing a well known
optimization (zero-copy I/O). Fujita Tomonori explained in the same
thread that it is
possible to improve the performance of STGT further, but that this would require
a lot of effort (implementing asynchronous I/O in the kernel and also
implementing
a new caching mechanism using pre-registered buffers).
See also:
http://sourceforge.net/mailarchive/forum.php?forum_name=scst-devel&viewmonth=200801&viewday=17
Bart Van Assche.
On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
>
> iSER has parameters to limit the maximum size of RDMA (it needs to
> repeat RDMA with a poor configuration)?
Please specify which parameters you are referring to. As you know I
had already repeated my tests with ridiculously high values for the
following iSER parameters: FirstBurstLength, MaxBurstLength and
MaxRecvDataSegmentLength (16 MB, which is more than the 1 MB block
size specified to dd).
Bart Van Assche.
On Wed, 30 Jan 2008 09:38:04 +0100
"Bart Van Assche" <[email protected]> wrote:
> On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
> >
> > iSER has parameters to limit the maximum size of RDMA (it needs to
> > repeat RDMA with a poor configuration)?
>
> Please specify which parameters you are referring to. As you know I
Sorry, I can't say. I don't know much about iSER. But seems that Pete
and Robin can get the better I/O performance - line speed ratio with
STGT.
The version of OpenIB might matters too. For example, Pete said that
STGT reads loses about 100 MB/s for some transfer sizes for some
transfer sizes due to the OpenIB version difference or other unclear
reasons.
http://article.gmane.org/gmane.linux.iscsi.tgt.devel/135
It's fair to say that it takes long time and need lots of knowledge to
get the maximum performance of SAN, I think.
I think that it would be easier to convince James with the detailed
analysis (e.g. where does it take so long, like Pete did), not just
'dd' performance results.
Pushing iSCSI target code into mainline failed four times: IET, SCST,
STGT (doing I/Os in kernel in the past), and PyX's one (*1). iSCSI
target code is huge. You said SCST comprises 14,000 lines, but it's
not iSCSI target code. The SCSI engine code comprises 14,000
lines. You need another 10,000 lines for the iSCSI driver. Note that
SCST's iSCSI driver provides only basic iSCSI features. PyX's iSCSI
target code implemenents more iSCSI features (like MC/S, ERL2, etc)
and comprises about 60,000 lines and it still lacks some features like
iSER, bidi, etc.
I think that it's reasonable to say that we need more than 'dd'
results before pushing about possible more than 60,000 lines to
mainline.
(*1) http://linux-iscsi.org/
James Bottomley wrote:
> The two target architectures perform essentially identical functions, so
> there's only really room for one in the kernel. Right at the moment,
> it's STGT. Problems in STGT come from the user<->kernel boundary which
> can be mitigated in a variety of ways. The fact that the figures are
> pretty much comparable on non IB networks shows this.
>
> I really need a whole lot more evidence than at worst a 20% performance
> difference on IB to pull one implementation out and replace it with
> another. Particularly as there's no real evidence that STGT can't be
> tweaked to recover the 20% even on IB.
James,
Although the performance difference between STGT and SCST is apparent,
this isn't the only point why SCST is better. I've already written about
it many times in various mailing lists, but let me summarize it one more
time here.
As you know, almost all kernel parts can be done in user space,
including all the drivers, networking, I/O management with block/SCSI
initiator subsystem and disk cache manager. But does it mean that
currently Linux kernel is bad and all the above should be (re)done in
user space instead? I believe, not. Linux isn't a microkernel for very
pragmatic reasons: simplicity and performance. So, additional important
point why SCST is better is simplicity.
For SCSI target, especially with hardware target card, data are came
from kernel and eventually served by kernel, which does actual I/O or
getting/putting data from/to cache. Dividing requests processing between
user and kernel space creates unnecessary interface layer(s) and
effectively makes the requests processing job distributed with all its
complexity and reliability problems. From my point of view, having such
distribution, where user space is master side and kernel is slave is
rather wrong, because:
1. It makes kernel depend from user program, which services it and
provides for it its routines, while the regular paradigm is the
opposite: kernel services user space applications. As a direct
consequence from it that there is no real protection for the kernel from
faults in the STGT core code without excessive effort, which, no
surprise, wasn't currently done and, seems, is never going to be done.
So, on practice debugging and developing under STGT isn't easier, than
if the whole code was in the kernel space, but, actually, harder (see
below why).
2. It requires new complicated interface between kernel and user spaces
that creates additional maintenance and debugging headaches, which don't
exist for kernel only code. Linus Torvalds some time ago perfectly
described why it is bad, see http://lkml.org/lkml/2007/4/24/451,
http://lkml.org/lkml/2006/7/1/41 and http://lkml.org/lkml/2007/4/24/364.
3. It makes for SCSI target impossible to use (at least, on a simple and
sane way) many effective optimizations: zero-copy cached I/O, more
control over read-ahead, device queue unplugging-plugging, etc. One
example of already implemented such features is zero-copy network data
transmission, done in simple 260 lines put_page_callback patch. This
optimization is especially important for the user space gate (scst_user
module), see below for details.
The whole point that development for kernel is harder, than for user
space, is totally nonsense nowadays. It's different, yes, in some ways
more limited, yes, but not harder. For ones who need gdb (I for many
years - don't) kernel has kgdb, plus it also has many not available for
user space or more limited there debug facilities like lockdep, lockup
detection, oprofile, etc. (I don't mention wider choice of more
effectively implemented synchronization primitives and not only them).
For people who need complicated target devices emulation, like, e.g., in
case of VTL (Virtual Tape Library), where there is a need to operate
with large mmap'ed memory areas, SCST provides gateway to the user space
(scst_user module), but, in contrast with STGT, it's done in regular
"kernel - master, user application - slave" paradigm, so it's reliable
and no fault in user space device emulator can break kernel and other
user space applications. Plus, since SCSI target state machine and
memory management are in the kernel, it's very effective and allows only
one kernel-user space switch per SCSI command.
Also, I should note here, that in the current state STGT in many aspects
doesn't fully conform SCSI specifications, especially in area of
management events, like Unit Attentions generation and processing, and
it doesn't look like somebody cares about it. At the same time, SCST
pays big attention to fully conform SCSI specifications, because price
of non-conformance is a possible user's data corruption.
Returning to performance, modern SCSI transports, e.g. InfiniBand, have
as low link latency as 1(!) microsecond. For comparison, the
inter-thread context switch time on a modern system is about the same,
syscall time - about 0.1 microsecond. So, only ten empty syscalls or one
context switch add the same latency as the link. Even 1Gbps Ethernet has
less, than 100 microseconds of round-trip latency.
You, probably, know, that QLogic Fibre Channel target driver for SCST
allows commands being executed either directly from soft IRQ, or from
the corresponding thread. There is a steady 5-7% difference in IOPS
between those modes on 512 bytes reads on nullio using 4Gbps link. So, a
single additional inter-kernel-thread context switch costs 5-7% of IOPS.
Another source of additional unavoidable with the user space approach
latency is data copy to/from cache. With the fully kernel space
approach, cache can be used directly, so no extra copy will be needed.
We can estimate how much latency the data copying adds. On the modern
systems memory copy throughput is less than 2GB/s, so on 20Gbps
InfiniBand link it almost doubles data transfer latency.
So, putting code in the user space you should accept the extra latency
it adds. Many, if not most, real-life workloads more or less latency,
not throughput, bound, so there shouldn't be surprise that single stream
"dd if=/dev/sdX of=/dev/null" on initiator gives too low values. Such
"benchmark" isn't less important and practical, than all the
multithreaded latency insensitive benchmarks, which people like running,
because it does essentially the same as most Linux processes do when
they read data from files.
You may object me that the target's backstorage device(s) latency is a
lot more, than 1 microsecond, but that is relevant only if data are
read/written from/to the actual backstorage media, not from the cache,
even from the backstorage device's cache. Nothing prevents target from
having 8 or even 64GB of cache, so most even random accesses could be
served by it. This is especially important for sync writes.
Thus, why SCST is better:
1. It is more simple, because it's monolithic, so all its components are
in one place and communicate using direct function calls. Hence, it is
smaller, faster, more reliable and maintainable. Currently it's bigger,
than STGT, just because it supports more features, see (2).
2. It supports more features: 1 to many pass-through support with all
necessary for it functionality, including support for non-disk SCSI
devices, like tapes, SGV cache, BLOCKIO, where requests converted to
bio's and directly sent to block level (this mode is effective for
random mostly workloads with data set size >> memory size on the
target), etc.
3. It has better performance and going to have it even better. SCST only
now enters in the phase, where it starts exploiting all advantages of
being in the kernel. Particularly, zero-copy cached I/O is currently
being implemented.
4. It provides safer and more effective interface to emulate target
devices in the user space via scst_user module.
5. It much more confirms to SCSI specifications (see above).
Vlad
FUJITA Tomonori wrote:
> On Tue, 29 Jan 2008 13:31:52 -0800
> Roland Dreier <[email protected]> wrote:
>
>
>> > . . STGT read SCST read . STGT read SCST read .
>> > . . performance performance . performance performance .
>> > . . (0.5K, MB/s) (0.5K, MB/s) . (1 MB, MB/s) (1 MB, MB/s) .
>> > . iSER (8 Gb/s network) . 250 N/A . 360 N/A .
>> > . SRP (8 Gb/s network) . N/A 421 . N/A 683 .
>>
>> > On the comparable figures, which only seem to be IPoIB they're showing a
>> > 13-18% variance, aren't they? Which isn't an incredible difference.
>>
>>Maybe I'm all wet, but I think iSER vs. SRP should be roughly
>>comparable. The exact formatting of various messages etc. is
>>different but the data path using RDMA is pretty much identical. So
>>the big difference between STGT iSER and SCST SRP hints at some big
>>difference in the efficiency of the two implementations.
>
>
> iSER has parameters to limit the maximum size of RDMA (it needs to
> repeat RDMA with a poor configuration)?
>
>
> Anyway, here's the results from Robin Humble:
>
> iSER to 7G ramfs, x86_64, centos4.6, 2.6.22 kernels, git tgtd,
> initiator end booted with mem=512M, target with 8G ram
>
> direct i/o dd
> write/read 800/751 MB/s
> dd if=/dev/zero of=/dev/sdc bs=1M count=5000 oflag=direct
> dd of=/dev/null if=/dev/sdc bs=1M count=5000 iflag=direct
>
> http://www.mail-archive.com/[email protected]/msg13502.html
>
> I think that STGT is pretty fast with the fast backing storage.
How fast SCST will be on the same hardware?
> I don't think that there is the notable perfornace difference between
> kernel-space and user-space SRP (or ISER) implementations about moving
> data between hosts. IB is expected to enable user-space applications
> to move data between hosts quickly (if not, what can IB provide us?).
>
> I think that the question is how fast user-space applications can do
> I/Os ccompared with I/Os in kernel space. STGT is eager for the advent
> of good asynchronous I/O and event notification interfances.
>
> One more possible optimization for STGT is zero-copy data
> transfer. STGT uses pre-registered buffers and move data between page
> cache and thsse buffers, and then does RDMA transfer. If we implement
> own caching mechanism to use pre-registered buffers directly with (AIO
> and O_DIRECT), then STGT can move data without data copies.
Great! So, you are going to duplicate Linux page cache in the user
space. You will continue keeping the in-kernel code as small as possible
and its mainteinership effort as low as possible by the cost that the
user space part's code size and complexity (and, hence, its
mainteinership effort) will rocket to the sky. Apparently, this doesn't
look like a good design decision.
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
FUJITA Tomonori wrote:
> On Wed, 30 Jan 2008 09:38:04 +0100
> "Bart Van Assche" <[email protected]> wrote:
>
>
>>On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
>>
>>>iSER has parameters to limit the maximum size of RDMA (it needs to
>>>repeat RDMA with a poor configuration)?
>>
>>Please specify which parameters you are referring to. As you know I
>
>
> Sorry, I can't say. I don't know much about iSER. But seems that Pete
> and Robin can get the better I/O performance - line speed ratio with
> STGT.
>
> The version of OpenIB might matters too. For example, Pete said that
> STGT reads loses about 100 MB/s for some transfer sizes for some
> transfer sizes due to the OpenIB version difference or other unclear
> reasons.
>
> http://article.gmane.org/gmane.linux.iscsi.tgt.devel/135
>
> It's fair to say that it takes long time and need lots of knowledge to
> get the maximum performance of SAN, I think.
>
> I think that it would be easier to convince James with the detailed
> analysis (e.g. where does it take so long, like Pete did), not just
> 'dd' performance results.
>
> Pushing iSCSI target code into mainline failed four times: IET, SCST,
> STGT (doing I/Os in kernel in the past), and PyX's one (*1). iSCSI
> target code is huge. You said SCST comprises 14,000 lines, but it's
> not iSCSI target code. The SCSI engine code comprises 14,000
> lines. You need another 10,000 lines for the iSCSI driver. Note that
> SCST's iSCSI driver provides only basic iSCSI features. PyX's iSCSI
> target code implemenents more iSCSI features (like MC/S, ERL2, etc)
> and comprises about 60,000 lines and it still lacks some features like
> iSER, bidi, etc.
>
> I think that it's reasonable to say that we need more than 'dd'
> results before pushing about possible more than 60,000 lines to
> mainline.
Tomo, please stop counting in-kernel lines only (see
http://lkml.org/lkml/2007/4/24/364). The amount of the overall project
lines for the same feature set is a lot more important.
> (*1) http://linux-iscsi.org/
>
On Jan 30, 2008 11:56 AM, FUJITA Tomonori <[email protected]> wrote:
> On Wed, 30 Jan 2008 09:38:04 +0100
> "Bart Van Assche" <[email protected]> wrote:
> >
> > Please specify which parameters you are referring to. As you know I
>
> Sorry, I can't say. I don't know much about iSER. But seems that Pete
> and Robin can get the better I/O performance - line speed ratio with
> STGT.
Robin Humble was using a DDR InfiniBand network, while my tests were
performed with an SDR InfiniBand network. Robin's results can't be
directly compared to my results.
Pete Wyckoff's results
(http://www.osc.edu/~pw/papers/wyckoff-iser-snapi07-talk.pdf) are hard
to interpret. I have asked Pete which of the numbers in his test can
be compared with what I measured, but Pete did not reply.
> The version of OpenIB might matters too. For example, Pete said that
> STGT reads loses about 100 MB/s for some transfer sizes for some
> transfer sizes due to the OpenIB version difference or other unclear
> reasons.
>
> http://article.gmane.org/gmane.linux.iscsi.tgt.devel/135
Pete wrote about a degradation from 600 MB/s to 500 MB/s for reads
with STGT+iSER. In my tests I measured 589 MB/s for reads (direct
I/O), which matches with the better result obtained by Pete.
Note: the InfiniBand kernel modules I used were those from the
2.6.22.9 kernel, not from the OFED distribution.
Bart.
On Wed, 30 Jan 2008 14:10:47 +0100
"Bart Van Assche" <[email protected]> wrote:
> On Jan 30, 2008 11:56 AM, FUJITA Tomonori <[email protected]> wrote:
> > On Wed, 30 Jan 2008 09:38:04 +0100
> > "Bart Van Assche" <[email protected]> wrote:
> > >
> > > Please specify which parameters you are referring to. As you know I
> >
> > Sorry, I can't say. I don't know much about iSER. But seems that Pete
> > and Robin can get the better I/O performance - line speed ratiwo with
> > STGT.
>
> Robin Humble was using a DDR InfiniBand network, while my tests were
> performed with an SDR InfiniBand network. Robin's results can't be
> directly compared to my results.
I know that you use different hardware. I used 'ratio' word.
BTW, you said the performance difference of dio READ is 38% but I
think it's 27.3 %, though it's still large.
> Pete Wyckoff's results
> (http://www.osc.edu/~pw/papers/wyckoff-iser-snapi07-talk.pdf) are hard
> to interpret. I have asked Pete which of the numbers in his test can
> be compared with what I measured, but Pete did not reply.
>
> > The version of OpenIB might matters too. For example, Pete said that
> > STGT reads loses about 100 MB/s for some transfer sizes for some
> > transfer sizes due to the OpenIB version difference or other unclear
> > reasons.
> >
> > http://article.gmane.org/gmane.linux.iscsi.tgt.devel/135
>
> Pete wrote about a degradation from 600 MB/s to 500 MB/s for reads
> with STGT+iSER. In my tests I measured 589 MB/s for reads (direct
> I/O), which matches with the better result obtained by Pete.
I don't know he used the same benchmark software so I don't think that
we can compare them.
All I tried to say is the OFED version might has big effect on the
performance. So you might need to find the best one.
> Note: the InfiniBand kernel modules I used were those from the
> 2.6.22.9 kernel, not from the OFED distribution.
I'm talking about a target machine (I think that Pete was also talking
about OFED on his target machine). STGT uses OFED libraries, I think.
On Wed, 2008-01-30 at 09:29 +0100, Bart Van Assche wrote:
> On Jan 29, 2008 9:42 PM, James Bottomley
> <[email protected]> wrote:
> > > As an SCST user, I would like to see the SCST kernel code integrated
> > > in the mainstream kernel because of its excellent performance on an
> > > InfiniBand network. Since the SCST project comprises about 14 KLOC,
> > > reviewing the SCST code will take considerable time. Who will do this
> > > reviewing work ? And with regard to the comments made by the
> > > reviewers: Vladislav, do you have the time to carry out the
> > > modifications requested by the reviewers ? I expect a.o. that
> > > reviewers will ask to move SCST's configuration pseudofiles from
> > > procfs to sysfs.
> >
> > The two target architectures perform essentially identical functions, so
> > there's only really room for one in the kernel. Right at the moment,
> > it's STGT. Problems in STGT come from the user<->kernel boundary which
> > can be mitigated in a variety of ways. The fact that the figures are
> > pretty much comparable on non IB networks shows this.
>
> Are you saying that users who need an efficient iSCSI implementation
> should switch to OpenSolaris ?
I'd certainly say that's a totally unsupported conclusion.
> The OpenSolaris COMSTAR project involves
> the migration of the existing OpenSolaris iSCSI target daemon from
> userspace to their kernel. The OpenSolaris developers are
> spending time on this because they expect a significant performance
> improvement.
Just because Solaris takes a particular design decision doesn't
automatically make it the right course of action.
Microsoft once pulled huge gobs of the C library and their windowing
system into the kernel in the name of efficiency. It proved not only to
be less efficient, but also to degrade their security model.
Deciding what lives in userspace and what should be in the kernel lies
at the very heart of architectural decisions. However, the argument
that "it should be in the kernel because that would make it faster" is
pretty much a discredited one. To prevail on that argument, you have to
demonstrate that there's no way to enable user space to do the same
thing at the same speed. Further, it was the same argument used the
last time around when the STGT vs SCST investigation was done. Your own
results on non-IB networks show that both architectures perform at the
same speed. That tends to support the conclusion that there's something
specific about IB that needs to be tweaked or improved for STGT to get
it to perform correctly.
Furthermore, if you have already decided before testing that SCST is
right and that STGT is wrong based on the architectures, it isn't
exactly going to increase my confidence in your measurement methodology
claiming to show this, now is it?
> > I really need a whole lot more evidence than at worst a 20% performance
> > difference on IB to pull one implementation out and replace it with
> > another. Particularly as there's no real evidence that STGT can't be
> > tweaked to recover the 20% even on IB.
>
> My measurements on a 1 GB/s InfiniBand network have shown that the current
> SCST implementation is able to read data via direct I/O at a rate of 811 GB/s
> (via SRP) and that the current STGT implementation is able to transfer data at a
> rate of 589 MB/s (via iSER). That's a performance difference of 38%.
>
> And even more important, the I/O latency of SCST is significantly
> lower than that
> of STGT. This is very important for database workloads -- the I/O pattern caused
> by database software is close to random I/O, and database software needs low
> latency I/O in order to run efficiently.
>
> In the thread with the title "Performance of SCST versus STGT" on the
> SCST-devel /
> STGT-devel mailing lists not only the raw performance numbers were discussed but
> also which further performance improvements are possible. It became clear that
> the SCST performance can be improved further by implementing a well known
> optimization (zero-copy I/O). Fujita Tomonori explained in the same
> thread that it is
> possible to improve the performance of STGT further, but that this would require
> a lot of effort (implementing asynchronous I/O in the kernel and also
> implementing
> a new caching mechanism using pre-registered buffers).
These are both features being independently worked on, are they not?
Even if they weren't, the combination of the size of SCST in kernel plus
the problem of having to find a migration path for the current STGT
users still looks to me to involve the greater amount of work.
James
On Wed, 2008-01-30 at 09:38 +0100, Bart Van Assche wrote:
> On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
> >
> > iSER has parameters to limit the maximum size of RDMA (it needs to
> > repeat RDMA with a poor configuration)?
>
> Please specify which parameters you are referring to. As you know I
> had already repeated my tests with ridiculously high values for the
> following iSER parameters: FirstBurstLength, MaxBurstLength and
> MaxRecvDataSegmentLength (16 MB, which is more than the 1 MB block
> size specified to dd).
the 1Mb block size is a bit of a red herring. Unless you've
specifically increased the max_sector_size and are using an sg_chain
converted driver, on x86 the maximum possible transfer accumulation is
0.5MB.
I certainly don't rule out that increasing the transfer size up from
0.5MB might be the way to improve STGT efficiency, since at an 1GB/s
theoretical peak, that's roughly 2000 context switches per I/O; however,
It doesn't look like you've done anything that will overcome the block
layer limitations.
James
On Jan 30, 2008 5:34 PM, James Bottomley
<[email protected]> wrote:
>
> On Wed, 2008-01-30 at 09:38 +0100, Bart Van Assche wrote:
> > On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
> > >
> > > iSER has parameters to limit the maximum size of RDMA (it needs to
> > > repeat RDMA with a poor configuration)?
> >
> > Please specify which parameters you are referring to. As you know I
> > had already repeated my tests with ridiculously high values for the
> > following iSER parameters: FirstBurstLength, MaxBurstLength and
> > MaxRecvDataSegmentLength (16 MB, which is more than the 1 MB block
> > size specified to dd).
>
> the 1Mb block size is a bit of a red herring. Unless you've
> specifically increased the max_sector_size and are using an sg_chain
> converted driver, on x86 the maximum possible transfer accumulation is
> 0.5MB.
I did not publish the results, but I have also done tests with other
block sizes. The other sizes I tested were between 0.1MB and 10MB. The
performance difference for these other sizes compared to a block size
of 1MB was small (smaller than the variance between individual tests
results).
Bart.
On Jan 30, 2008 5:22 PM, James Bottomley
<[email protected]> wrote:
> ...
> Deciding what lives in userspace and what should be in the kernel lies
> at the very heart of architectural decisions. However, the argument
> that "it should be in the kernel because that would make it faster" is
> pretty much a discredited one. To prevail on that argument, you have to
> demonstrate that there's no way to enable user space to do the same
> thing at the same speed. Further, it was the same argument used the
> last time around when the STGT vs SCST investigation was done. Your own
> results on non-IB networks show that both architectures perform at the
> same speed. That tends to support the conclusion that there's something
> specific about IB that needs to be tweaked or improved for STGT to get
> it to perform correctly.
You should know that given two different implementations in software of the
same communication protocol, differences in latency and throughput become
more visible as the network latency gets lower and the throughput gets higher.
That's why conclusions can only be drawn from the InfiniBand numbers, and
not from the 1 Gbit/s Ethernet numbers. Assuming that there is something
specific in STGT with regard to InfiniBand is speculation.
> Furthermore, if you have already decided before testing that SCST is
> right and that STGT is wrong based on the architectures, it isn't
> exactly going to increase my confidence in your measurement methodology
> claiming to show this, now is it?
I did not draw any conclusions from the architecture -- the only data I based my
conclusions on were my own performance measurements.
> ...
> These are both features being independently worked on, are they not?
> Even if they weren't, the combination of the size of SCST in kernel plus
> the problem of having to find a migration path for the current STGT
> users still looks to me to involve the greater amount of work.
My proposal was to have both the SCST kernel code and the STGT kernel
code in the mainstream Linux kernel. This would make it easier for current
STGT users to evaluate SCST. It's too early to choose one of the two
projects -- this choice can be made later on.
Bart.
On Jan 30, 2008 2:54 PM, FUJITA Tomonori <[email protected]> wrote:
> On Wed, 30 Jan 2008 14:10:47 +0100
> "Bart Van Assche" <[email protected]> wrote:
>
> > On Jan 30, 2008 11:56 AM, FUJITA Tomonori <[email protected]> wrote:
> > >
> > > Sorry, I can't say. I don't know much about iSER. But seems that Pete
> > > and Robin can get the better I/O performance - line speed ratio with
> > > STGT.
> >
> > Robin Humble was using a DDR InfiniBand network, while my tests were
> > performed with an SDR InfiniBand network. Robin's results can't be
> > directly compared to my results.
>
> I know that you use different hardware. I used 'ratio' word.
Let's start with summarizing the relevant numbers from Robin's
measurements and my own measurements.
Maximum bandwidth of the underlying physical medium: 2000 MB/s for a
DDR 4x InfiniBand network and 1000 MB/s for a SDR 4x InfiniBand
network.
Maximum bandwidth reported by the OFED ib_write_bw test program: 1473
MB/s for Robin's setup and 933 MB/s for my setup. These numbers match
published ib_write_bw results (see e.g. figure 11 in
http://www.usenix.org/events/usenix06/tech/full_papers/liu/liu_html/index.html
or chapter 7 in
http://www.voltaire.com/ftp/rocks/HCA-4X0_Linux_GridStack_4.3_Release_Notes_DOC-00171-A00.pdf)
Throughput measured for communication via STGT + iSER to a remote RAM
disk via direct I/O with dd: 800 MB/s for writing and 751 MB/s for
reading in Robin's setup, and 647 MB/s for writing and 589 MB/s for
reading in my setup.
>From this we can compute the I/O-performance to ib_write_bw bandwidth:
54 % for writing and 51 % for reading in Robin's setup, and 69 % for
writing and 63 % for reading in my setup. Or a slightly better
utilization of the bandwidth in my setup than in Robin's setup. This
is no surprise -- the faster a communication link is, the harder it is
to use all of the available bandwidth.
So why did you state that in Robin's tests the I/O performance to line
speed ratio was better than in my tests ?
Bart Van Assche.
Greetings all,
On Wed, 2008-01-30 at 19:56 +0900, FUJITA Tomonori wrote:
> On Wed, 30 Jan 2008 09:38:04 +0100
> "Bart Van Assche" <[email protected]> wrote:
>
> > On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
> > >
> > > iSER has parameters to limit the maximum size of RDMA (it needs to
> > > repeat RDMA with a poor configuration)?
> >
> > Please specify which parameters you are referring to. As you know I
>
> Sorry, I can't say. I don't know much about iSER. But seems that Pete
> and Robin can get the better I/O performance - line speed ratio with
> STGT.
>
> The version of OpenIB might matters too. For example, Pete said that
> STGT reads loses about 100 MB/s for some transfer sizes for some
> transfer sizes due to the OpenIB version difference or other unclear
> reasons.
>
> http://article.gmane.org/gmane.linux.iscsi.tgt.devel/135
>
> It's fair to say that it takes long time and need lots of knowledge to
> get the maximum performance of SAN, I think.
>
> I think that it would be easier to convince James with the detailed
> analysis (e.g. where does it take so long, like Pete did), not just
> 'dd' performance results.
>
> Pushing iSCSI target code into mainline failed four times: IET, SCST,
> STGT (doing I/Os in kernel in the past), and PyX's one (*1). iSCSI
> target code is huge. You said SCST comprises 14,000 lines, but it's
> not iSCSI target code. The SCSI engine code comprises 14,000
> lines. You need another 10,000 lines for the iSCSI driver. Note that
> SCST's iSCSI driver provides only basic iSCSI features. PyX's iSCSI
> target code implemenents more iSCSI features (like MC/S, ERL2, etc)
> and comprises about 60,000 lines and it still lacks some features like
> iSER, bidi, etc.
>
The PyX storage engine supports a scatterlist linked list algorithm that
maps any sector count + sector size combination down to contiguous
struct scatterlist arrays across (potentially) multiple Linux storage
subsystems from a single CDB received on a initiator port. This design
was a consequence of a requirement for running said engine on Linux v2.2
and v2.4 across non cache coherent systems (MIPS R5900-EE) using a
single contiguous memory block mapped into struct buffer_head for PATA
access, and struct scsi_cmnd access on USB storage. Note that this was
before struct bio and struct scsi_request existed..
The PyX storage engine as it exists at Linux-iSCSI.org today can be
thought of as a hybrid OSD processing engine, as it maps storage object
memory across a number of tasks from a received command CDB. The
ability to pass in pre allocated memory from an RDMA capable adapter, as
well as allocated internally (ie: traditional iSCSI without open_iscsi's
struct skbuff rx zero-copy) is inherient in the design of the storage
engine. The lacking Bidi support can be attributed to lack of greater
support (and hence user interest) in Bidi, but I am really glad to see
this getting into the SCSI ML and STGT, and is certainly of interest in
the long term. Another feature that is missing in the current engine is
> 16 Byte CDBs, which I would imagine alot of vendor folks would like to
see in Linux as well. This is pretty easy to add in iSCSI with an AHS
and in the engine and storage subsystems.
> I think that it's reasonable to say that we need more than 'dd'
> results before pushing about possible more than 60,000 lines to
> mainline.
>
> (*1) http://linux-iscsi.org/
> -
The 60k lines of code also includes functionality (the SE mirroring
comes to mind) that I do not plan to push towards mainline, along with
other legacy bits so we can build on earlier v2.6 embedded platforms.
The existing Target mode LIO-SE that provides linked list scatterlist
mapping algorithm that is similar to what Jens and Rusty have been
working on, and is under 14k lines including the switch(cdb[0]) +
function pointer assignment to per CDB specific structure that is called
potentially out-of-order in the RX side context of the CmdSN state
machine in RFC-3720. The current SE is also lacking the very SCSI
specific task management state machines that not a whole lot of iSCSI
implementions implement properly, and seem to be minimal interest to
users, and of moderate interest to vendors. Getting this implemented
generically in SCSI, as opposed to an transport specific mechanisim
would benefit the Linux SCSI target engine.
The pSCSI (struct scsi_cmnd), iBlock (struct bio) and FILE (struct file)
plugins together are a grand total of 3.5k lines using the v2.9 LIO-SE
interface. Assuming we have a single preferred data and control patch
for underlying physical and virtual block devices, this could also get
smaller. A quick check of the code puts the traditional kernel level
iSCSI statemachine at roughly 16k, which is pretty good for the complete
state machine. Also, having iSER and traditional iSCSI share MC/S and
ERL=2 common code will be of interest, as well as iSCSI login state
machines, which are identical minus the extra iSER specific keys and
requirement to transition from byte stream mode to RDMA accelerated
mode.
Since this particular code is located in a non-data path critical
section, the kernel vs. user discussion is a wash. If we are talking
about data path, yes, the relevance of DD tests in kernel designs are
suspect :p. For those IB testers who are interested, perhaps having a
look with disktest from the Linux Test Project would give a better
comparision between the two implementations on a RDMA capable fabric
like IB for best case performance. I think everyone is interested in
seeing just how much data path overhead exists between userspace and
kernel space in typical and heavy workloads, if if this overhead can be
minimized to make userspace a better option for some of this very
complex code.
--nab
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
> Since this particular code is located in a non-data path critical
> section, the kernel vs. user discussion is a wash. If we are talking
> about data path, yes, the relevance of DD tests in kernel designs are
> suspect :p. For those IB testers who are interested, perhaps having a
> look with disktest from the Linux Test Project would give a better
> comparision between the two implementations on a RDMA capable fabric
> like IB for best case performance. I think everyone is interested in
> seeing just how much data path overhead exists between userspace and
> kernel space in typical and heavy workloads, if if this overhead can be
> minimized to make userspace a better option for some of this very
> complex code.
I can run disktest on the same setups I ran dd on. This will take some
time however.
Disktest is new to me -- any hints with regard to suitable
combinations of command line parameters are welcome. The most recent
version I could find on http://ltp.sourceforge.net/ is ltp-20071231.
Bart Van Assche.
Hi Bart,
On Thu, 2008-01-31 at 15:34 +0100, Bart Van Assche wrote:
> On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
> > Since this particular code is located in a non-data path critical
> > section, the kernel vs. user discussion is a wash. If we are talking
> > about data path, yes, the relevance of DD tests in kernel designs are
> > suspect :p. For those IB testers who are interested, perhaps having a
> > look with disktest from the Linux Test Project would give a better
> > comparision between the two implementations on a RDMA capable fabric
> > like IB for best case performance. I think everyone is interested in
> > seeing just how much data path overhead exists between userspace and
> > kernel space in typical and heavy workloads, if if this overhead can be
> > minimized to make userspace a better option for some of this very
> > complex code.
>
> I can run disktest on the same setups I ran dd on. This will take some
> time however.
>
> Disktest is new to me -- any hints with regard to suitable
> combinations of command line parameters are welcome. The most recent
> version I could find on http://ltp.sourceforge.net/ is ltp-20071231.
>
I posted some numbers with traditional iSCSI on Neterion Xframe I 10
Gb/sec with LRO back in 2005 with disktest on the 1st generation x86_64
hardware available at the time. These tests where designed to show the
performance advantages of internexus multiplexing that is available
within traditional iSCSI, as well as iSER.
The disktest parameters that I used are listed in the following thread:
https://www.redhat.com/archives/dm-devel/2005-April/msg00013.html
--nab
> Bart Van Assche.
>
Bart Van Assche wrote:
> On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
>
>>Since this particular code is located in a non-data path critical
>>section, the kernel vs. user discussion is a wash. If we are talking
>>about data path, yes, the relevance of DD tests in kernel designs are
>>suspect :p. For those IB testers who are interested, perhaps having a
>>look with disktest from the Linux Test Project would give a better
>>comparision between the two implementations on a RDMA capable fabric
>>like IB for best case performance. I think everyone is interested in
>>seeing just how much data path overhead exists between userspace and
>>kernel space in typical and heavy workloads, if if this overhead can be
>>minimized to make userspace a better option for some of this very
>>complex code.
>
> I can run disktest on the same setups I ran dd on. This will take some
> time however.
Disktest was already referenced in the beginning of the performance
comparison thread, but its results are not very interesting if we are
going to find out, which implementation is more effective, because in
the modes, in which usually people run this utility, it produces latency
insensitive workload (multiple threads working in parallel). So, such
multithreaded disktests results will be different between STGT and SCST
only if STGT's implementation will get target CPU bound. If CPU on the
target is powerful enough, even extra busy loops in the STGT or SCST hot
path code will change nothing.
Additionally, multithreaded disktest over RAM disk is a good example of
a synthetic benchmark, which has almost no relation with real life
workloads. But people like it, because it produces nice looking results.
Actually, I don't know what kind of conclusions it is possible to make
from disktest's results (maybe only how throughput gets bigger or slower
with increasing number of threads?), it's a good stress test tool, but
not more.
> Disktest is new to me -- any hints with regard to suitable
> combinations of command line parameters are welcome. The most recent
> version I could find on http://ltp.sourceforge.net/ is ltp-20071231.
>
> Bart Van Assche.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
Vladislav Bolkhovitin wrote:
> Bart Van Assche wrote:
[...]
>> I can run disktest on the same setups I ran dd on. This will take some
>> time however.
>
> Disktest was already referenced in the beginning of the performance
> comparison thread, but its results are not very interesting if we are
> going to find out, which implementation is more effective, because in
> the modes, in which usually people run this utility, it produces latency
> insensitive workload (multiple threads working in parallel). So, such
There are other issues with disktest, in that you can easily specify
option combinations that generate apparently 5+ GB/s of IO, though
actual traffic over the link to storage is very low. Caveat disktest
emptor.
> multithreaded disktests results will be different between STGT and SCST
> only if STGT's implementation will get target CPU bound. If CPU on the
> target is powerful enough, even extra busy loops in the STGT or SCST hot
> path code will change nothing.
>
> Additionally, multithreaded disktest over RAM disk is a good example of
> a synthetic benchmark, which has almost no relation with real life
> workloads. But people like it, because it produces nice looking results.
I agree. The backing store should be a disk for it to have meaning,
though please note my caveat above.
>
> Actually, I don't know what kind of conclusions it is possible to make
> from disktest's results (maybe only how throughput gets bigger or slower
> with increasing number of threads?), it's a good stress test tool, but
> not more.
Unfortunately, I agree. Bonnie++, dd tests, and a few others seem to
bear far closer to "real world" tests than disktest and iozone, the
latter of which does more to test the speed of RAM cache and system call
performance than actual IO.
>> Disktest is new to me -- any hints with regard to suitable
>> combinations of command line parameters are welcome. The most recent
>> version I could find on http://ltp.sourceforge.net/ is ltp-20071231.
>>
>> Bart Van Assche.
Here is what I have run:
disktest -K 8 -B 256k -I F -N 20000000 -P A -w /big/file
disktest -K 8 -B 64k -I F -N 20000000 -P A -w /big/file
disktest -K 8 -B 1k -I B -N 2000000 -P A /dev/sdb2
and many others.
Joe
--
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: [email protected]
web : http://www.scalableinformatics.com
http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423
fax : +1 866 888 3112
cell : +1 734 612 4615
On Jan 31, 2008 5:25 PM, Joe Landman <[email protected]> wrote:
> Vladislav Bolkhovitin wrote:
> > Actually, I don't know what kind of conclusions it is possible to make
> > from disktest's results (maybe only how throughput gets bigger or slower
> > with increasing number of threads?), it's a good stress test tool, but
> > not more.
>
> Unfortunately, I agree. Bonnie++, dd tests, and a few others seem to
> bear far closer to "real world" tests than disktest and iozone, the
> latter of which does more to test the speed of RAM cache and system call
> performance than actual IO.
I have ran some tests with Bonnie++, but found out that on a fast
network like IB the filesystem used for the test has a really big
impact on the test results.
If anyone has a suggestion for a better test than dd to compare the
performance of SCSI storage protocols, please let it know.
Bart Van Assche.
Bart Van Assche wrote:
> I have ran some tests with Bonnie++, but found out that on a fast
> network like IB the filesystem used for the test has a really big
> impact on the test results.
This is true of the file systems when physically directly connected to
the unit as well. Some file systems are designed with high performance
in mind, some are not.
> If anyone has a suggestion for a better test than dd to compare the
> performance of SCSI storage protocols, please let it know.
Hmmm... if you care about the protocol side, I can't help. Our users
are more concerned with the file system side, so this is where we focus
our tuning attention.
>
> Bart Van Assche.
Joe
--
Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: [email protected]
web : http://www.scalableinformatics.com
http://jackrabbit.scalableinformatics.com
phone: +1 734 786 8423
fax : +1 866 888 3112
cell : +1 734 612 4615
On Thu, 2008-01-31 at 18:50 +0300, Vladislav Bolkhovitin wrote:
> Bart Van Assche wrote:
> > On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
> >
> >>Since this particular code is located in a non-data path critical
> >>section, the kernel vs. user discussion is a wash. If we are talking
> >>about data path, yes, the relevance of DD tests in kernel designs are
> >>suspect :p. For those IB testers who are interested, perhaps having a
> >>look with disktest from the Linux Test Project would give a better
> >>comparision between the two implementations on a RDMA capable fabric
> >>like IB for best case performance. I think everyone is interested in
> >>seeing just how much data path overhead exists between userspace and
> >>kernel space in typical and heavy workloads, if if this overhead can be
> >>minimized to make userspace a better option for some of this very
> >>complex code.
> >
> > I can run disktest on the same setups I ran dd on. This will take some
> > time however.
>
> Disktest was already referenced in the beginning of the performance
> comparison thread, but its results are not very interesting if we are
> going to find out, which implementation is more effective, because in
> the modes, in which usually people run this utility, it produces latency
> insensitive workload (multiple threads working in parallel). So, such
> multithreaded disktests results will be different between STGT and SCST
> only if STGT's implementation will get target CPU bound. If CPU on the
> target is powerful enough, even extra busy loops in the STGT or SCST hot
> path code will change nothing.
>
I think the really interesting numbers are the difference for bulk I/O
between kernel and userspace on both traditional iSCSI and the RDMA
enabled flavours. I have not been able to determine anything earth
shattering from the current run of kernel vs. userspace tests, nor which
method of implementation for iSER, SRP, and generic Storage Engine are
'more effective' for that case. Performance and latency to real storage
would make alot more sense for the kernel vs. user case. Also workloads
against software LVM and Linux MD block devices would be of interest as
these would be some of the more typical deployments that would be in the
field, and is what Linux-iSCSI.org uses for our production cluster
storage today.
Having implemented my own iSCSI and SCSI Target mode Storage Engine
leads me to believe that putting logic in userspace is probably a good
idea in the longterm. If this means putting the entire data IO path
into userspace for Linux/iSCSI, then there needs to be a good reason why
this will not not scale to multi-port 10 Gb/sec engines in traditional
and RDMA mode if we need to take this codepath back into the kernel.
The end goal is to have the most polished and complete storage engine
and iSCSI stacks designs go upstream, which is something I think we can
all agree on.
Also, with STGT being a pretty new design which has not undergone alot
of optimization, perhaps profiling both pieces of code against similar
tests would give us a better idea of where userspace bottlenecks reside.
Also, the overhead involved with traditional iSCSI for bulk IO from
kernel / userspace would also be a key concern for a much larger set of
users, as iSER and SRP on IB is a pretty small userbase and will
probably remain small for the near future.
> Additionally, multithreaded disktest over RAM disk is a good example of
> a synthetic benchmark, which has almost no relation with real life
> workloads. But people like it, because it produces nice looking results.
>
Yes, people like to claim their stacks are the fastest with RAM disk
benchmarks. But hooking up their fast network silicon to existing
storage hardware and OS storage subsystems and software is where the
real game is..
> Actually, I don't know what kind of conclusions it is possible to make
> from disktest's results (maybe only how throughput gets bigger or slower
> with increasing number of threads?), it's a good stress test tool,
> but
> not more.
>
Being able to have a best case baseline with disktest for kernel vs.
user would be of interest for both transport protocol and SCSI Target
mode Storage Engine profiling. The first run of tests looked pretty
bandwith oriented, so disktest works well to determine maximum bandwith.
Disktest also is nice for getting reads from cache on hardware RAID
controllers because disktest only generates requests with LBAs from 0 ->
disktest BLOCKSIZE.
--nab
> > Disktest is new to me -- any hints with regard to suitable
> > combinations of command line parameters are welcome. The most recent
> > version I could find on http://ltp.sourceforge.net/ is ltp-20071231.
> >
> > Bart Van Assche.
> > -
> > To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> > the body of a message to [email protected]
> > More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
>
>
On Jan 31, 2008 6:14 PM, Nicholas A. Bellinger <[email protected]> wrote:
> Also, with STGT being a pretty new design which has not undergone alot
> of optimization, perhaps profiling both pieces of code against similar
> tests would give us a better idea of where userspace bottlenecks reside.
> Also, the overhead involved with traditional iSCSI for bulk IO from
> kernel / userspace would also be a key concern for a much larger set of
> users, as iSER and SRP on IB is a pretty small userbase and will
> probably remain small for the near future.
Two important trends in data center technology are server
consolidation and storage consolidation. A.o. every web hosting
company can profit from a fast storage solution. I wouldn't call this
a small user base.
Regarding iSER and SRP on IB: InfiniBand is today the most economic
solution for a fast storage network. I do not know which technology
will be the most popular for storage consolidation within a few years
-- this can be SRP, iSER, IPoIB + SDP, FCoE (Fibre Channel over
Ethernet) or maybe yet another technology. No matter which technology
becomes the most popular for storage applications, there will be a
need for high-performance storage software.
References:
* Michael Feldman, Battle of the Network Fabrics, HPCwire, December
2006, http://www.hpcwire.com/hpc/1145060.html
* NetApp, Reducing Data Center Power Consumption Through Efficient
Storage, February 2007,
http://www.netapp.com/ftp/wp-reducing-datacenter-power-consumption.pdf
Bart Van Assche.
On Thu, 2008-01-31 at 18:08 +0100, Bart Van Assche wrote:
> If anyone has a suggestion for a better test than dd to compare the
> performance of SCSI storage protocols, please let it know.
xdd on /dev/sda, sdb, etc. using -dio to do direct IO seems to work
decently, though it is hard (ie, impossible) to get a repeatable
sequence of IO when using higher queue depths, as it uses threads to
generate multiple requests.
You may also look at sgpdd_survey from Lustre's iokit, but I've not done
much with that -- it uses the sg devices to send lowlevel SCSI commands.
I've been playing around with some benchmark code using libaio, but it's
not in generally usable shape.
xdd:
http://www.ioperformance.com/products.htm
Lustre IO Kit:
http://manual.lustre.org/manual/LustreManual16_HTML/DynamicHTML-20-1.html
--
Dave Dillow
National Center for Computational Science
Oak Ridge National Laboratory
(865) 241-6602 office
On Thu, 2008-01-31 at 18:40 +0100, Bart Van Assche wrote:
> On Jan 31, 2008 6:14 PM, Nicholas A. Bellinger <[email protected]> wrote:
> > Also, with STGT being a pretty new design which has not undergone alot
> > of optimization, perhaps profiling both pieces of code against similar
> > tests would give us a better idea of where userspace bottlenecks reside.
> > Also, the overhead involved with traditional iSCSI for bulk IO from
> > kernel / userspace would also be a key concern for a much larger set of
> > users, as iSER and SRP on IB is a pretty small userbase and will
> > probably remain small for the near future.
>
> Two important trends in data center technology are server
> consolidation and storage consolidation. A.o. every web hosting
> company can profit from a fast storage solution. I wouldn't call this
> a small user base.
>
> Regarding iSER and SRP on IB: InfiniBand is today the most economic
> solution for a fast storage network. I do not know which technology
> will be the most popular for storage consolidation within a few years
> -- this can be SRP, iSER, IPoIB + SDP, FCoE (Fibre Channel over
> Ethernet) or maybe yet another technology. No matter which technology
> becomes the most popular for storage applications, there will be a
> need for high-performance storage software.
>
I meant small referring to storage on IB fabrics which has usually been
in the research and national lab settings, with some other vendors
offering IB as an alternative storage fabric for those who [w,c]ould not
wait for 10 Gb/sec copper Ethernet and Direct Data Placement to come
online. These types of numbers compared to say traditional iSCSI, that
is getting used all over the place these days in areas I won't bother
listing here.
As for the future, I am obviously cheering for IP storage fabrics, in
particular 10 Gb/sec Ethernet and Direct Data Placement in concert with
iSCSI Extentions for RDMA to give the data center a high performance,
low latency transport that can do OS independent storage multiplexing
and recovery across multiple independently developed implementions.
Also avoiding lock-in from un-interoptable storage transports (espically
on the high end) that had plauged so many vendors in years past has
become an real option in the past few years with a IETF defined block
level storage protocol. We are actually going on four years since
RFC-3720 was ratified. (April 2004)
Making the 'enterprise' ethernet switching equipment go from millisecond
to nanosecond latency in a whole different story that goes beyond my
area of expertise. I know there is one startup (Fulcrum Micro) who is
working on this problem and seems to be making some good progress.
--nab
> References:
> * Michael Feldman, Battle of the Network Fabrics, HPCwire, December
> 2006, http://www.hpcwire.com/hpc/1145060.html
> * NetApp, Reducing Data Center Power Consumption Through Efficient
> Storage, February 2007,
> http://www.netapp.com/ftp/wp-reducing-datacenter-power-consumption.pdf
>
> Bart Van Assche.
>
On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
>
> The PyX storage engine supports a scatterlist linked list algorithm that
> ...
Which parts of the PyX source code are licensed under the GPL and
which parts are closed source ? A Google query for PyX + iSCSI showed
information about licensing deals. Licensing deals can only be closed
for software that is not entirely licensed under the GPL.
Bart Van Assche.
On Jan 31, 2008 7:15 PM, Nicholas A. Bellinger <[email protected]> wrote:
>
> I meant small referring to storage on IB fabrics which has usually been
> in the research and national lab settings, with some other vendors
> offering IB as an alternative storage fabric for those who [w,c]ould not
> wait for 10 Gb/sec copper Ethernet and Direct Data Placement to come
> online. These types of numbers compared to say traditional iSCSI, that
> is getting used all over the place these days in areas I won't bother
> listing here.
InfiniBand has several advantages over 10 Gbit/s Ethernet (the list
below probably isn't complete):
- Lower latency. Communication latency is not only determined by the
latency of a switch. The whole InfiniBand protocol stack was designed
with low latency in mind. Low latency is really important for database
software that accesses storage over a network.
- High-availability is implemented at the network layer. Suppose that
a group of servers has dual-port network interfaces and is
interconnected via a so-called dual star topology, With an InfiniBand
network, failover in case of a single failure (link or switch) is
handled without any operating system or application intervention. With
Ethernet, failover in case of a single failure must be handled either
by the operating system or by the application.
- You do not have to use iSER or SRP to use the bandwidth of an
InfiniBand network effectively. The SDP (Sockets Direct Protocol)
makes it possible that applications benefit from RDMA by using the
very classic IPv4 Berkeley sockets interface. An SDP implementation in
software is already available today via OFED. iperf reports 470 MB/s
on single-threaded tests and 975 MB/s for a performance test with two
threads on an SDR 4x InfiniBand network. These tests were performed
with the OFED 1.2.5.4 SDP implementation. It is possible that future
SDP implementations will perform even better. (Note: I could not yet
get iSCSI over SDP working.)
We should leave the choice of networking technology open -- both
Ethernet and InfiniBand have specific advantages.
See also:
InfiniBand Trade Association, InfiniBand Architecture Specification
Release 1.2.1, http://www.infinibandta.org/specs/register/publicspec/
Bart Van Assche.
On Fri, 2008-02-01 at 09:11 +0100, Bart Van Assche wrote:
> On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
> >
> > The PyX storage engine supports a scatterlist linked list algorithm that
> > ...
>
> Which parts of the PyX source code are licensed under the GPL and
> which parts are closed source ? A Google query for PyX + iSCSI showed
> information about licensing deals. Licensing deals can only be closed
> for software that is not entirely licensed under the GPL.
>
I was using the name PyX to give an historical context to the
discussion. :-) In more recent times, I have been using the name "LIO
Target Stack" and "LIO Storage Engine" to refer to Traditional RFC-3720
Target statemachines, and SCSI Processing engine implementation
respectively. The codebase has matured significantly from the original
codebase, as the Linux SCSI, ATA and Block subsystems envolved from
v2.2, v2.4, v2.5 and modern v2.6, the LIO stack has grown (and sometimes
shrunk) along with the following requirement; To support all possible
storage devices on all subsystems on any hardware platform that Linux
could be made to boot. Interopt with other non Linux SCSI subsystems
was also an issue early in development.. If you can imagine a Solaris
SCSI subsystem asking for T10 EVPD WWN information from a Linux/iSCSI
Target with pre libata SATA drivers, you can probably guess just how
time was spent looking at packet captures to figure out to make OS
dependent (ie: outernexus) multipath to play nice.
Note that PyX Target Code for Linux v2.6 has been available in source
and binary form for a diverse array of Linux devices and environments
since September 2007. Right around this time, the Linux-iSCSI.org
Storage and Virtualization stack went online for the first time using
OCFS2, PVM, HVM, LVM, RAID6 and of course, traditional RFC-3720 on 10
Gb/sec and 1 Gb/sec fabric. There have also been world's first storage
research work and prototypes that have been developed with the LIO code.
Information on these topics is available from the homepage, and a few
links deep there are older projects and information about features
inherent to the LIO Target and Storage Engine. One of my items for the
v2.9 codebase in 2008 is start picking apart the current code and
determining which pieces should be sent upstream for review. I have
also been spending alot of time recently looking at the other available
open source storage transport and processing stacks and seeing how
Linux/iSCSI, and other projects can benefit from our large pool of
people, knowledge, and code.
Speaking of the LIO Target and SE code, it today runs the production
services for Linux-iSCSI.org and it's storage and virtualization
clusters on x86_64. It also also provides a base for next generation
and forward looking projects that exist (or soon to exist :-) within the
Linux/iSCSI ecosystem. There have been lots of time and resources put
into the codebase, and having a real live working RFC-3720 stack that
supports optional features that give iSCSI (and hence designed into
iSER) the flexibility and transparentness to operate as the original
designers intended.
Many thanks for your most valuable of time,
--nab
> Bart Van Assche.
>
On Feb 1, 2008 11:39 AM, Nicholas A. Bellinger <[email protected]> wrote:
>
> On Fri, 2008-02-01 at 09:11 +0100, Bart Van Assche wrote:
> > On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
> > >
> > > The PyX storage engine supports a scatterlist linked list algorithm that
> > > ...
> >
> > Which parts of the PyX source code are licensed under the GPL and
> > which parts are closed source ? A Google query for PyX + iSCSI showed
> > information about licensing deals. Licensing deals can only be closed
> > for software that is not entirely licensed under the GPL.
> >
>
> I was using the name PyX to give an historical context to the
> discussion. ...
Regarding the PyX Target Code: I have found a link via which I can
download a free 30-day demo. This means that a company is earning
money via this target code and that the source code is not licensed
under the GPL. This is fine, but it also means that today the PyX
target code is not a candidate for inclusion in the Linux kernel, and
that it is unlikely that all of the PyX target code (kernelspace +
userspace) will be made available under GPL soon.
Bart Van Assche.
Bart Van Assche wrote:
> On Jan 31, 2008 5:25 PM, Joe Landman <[email protected]> wrote:
>
>>Vladislav Bolkhovitin wrote:
>>
>>>Actually, I don't know what kind of conclusions it is possible to make
>>>from disktest's results (maybe only how throughput gets bigger or slower
>>>with increasing number of threads?), it's a good stress test tool, but
>>>not more.
>>
>>Unfortunately, I agree. Bonnie++, dd tests, and a few others seem to
>>bear far closer to "real world" tests than disktest and iozone, the
>>latter of which does more to test the speed of RAM cache and system call
>>performance than actual IO.
>
>
> I have ran some tests with Bonnie++, but found out that on a fast
> network like IB the filesystem used for the test has a really big
> impact on the test results.
>
> If anyone has a suggestion for a better test than dd to compare the
> performance of SCSI storage protocols, please let it know.
I would suggest you to try something from real life, like:
- Copying large file tree over a single or multiple IB links
- Measure of some DB engine's TPC
- etc.
> Bart Van Assche.
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Scst-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scst-devel
>
On Fri, 2008-02-01 at 12:04 +0100, Bart Van Assche wrote:
> On Feb 1, 2008 11:39 AM, Nicholas A. Bellinger <[email protected]> wrote:
> >
> > On Fri, 2008-02-01 at 09:11 +0100, Bart Van Assche wrote:
> > > On Jan 31, 2008 2:25 PM, Nicholas A. Bellinger <[email protected]> wrote:
> > > >
> > > > The PyX storage engine supports a scatterlist linked list algorithm that
> > > > ...
> > >
> > > Which parts of the PyX source code are licensed under the GPL and
> > > which parts are closed source ? A Google query for PyX + iSCSI showed
> > > information about licensing deals. Licensing deals can only be closed
> > > for software that is not entirely licensed under the GPL.
> > >
> >
> > I was using the name PyX to give an historical context to the
> > discussion. ...
>
> Regarding the PyX Target Code: I have found a link via which I can
> download a free 30-day demo. This means that a company is earning
> money via this target code and that the source code is not licensed
> under the GPL. This is fine, but it also means that today the PyX
> target code is not a candidate for inclusion in the Linux kernel, and
> that it is unlikely that all of the PyX target code (kernelspace +
> userspace) will be made available under GPL soon.
>
All of the kernel and C userspace code is open source and available from
linux-iscsi.org and licensed under the GPL. There is the BSD licensed
code from userspace (iSNS), as well as ISCSI and SCSI MIBs. As for what
pieces of code will be going upstream (for kernel and/or userspace), LIO
Target state machines and SE algoritims are definately some of the best
examples of GPL code for production IP storage fabric and has gained
maturity from people and resources applied to it in a number of
respects.
The LIO stack presents a number of possible options to get the diverse
amount of hardware and software to work. Completely dismissing the
available code is certainly a waste, and there are still significant
amounts of functionality related to real-time administration, RFC-3720
MC/S and ERL=2 and generic SE functionality OS storage subsystems that
only exist in LIO and our assoicated projects. A one obvious example is
the LIO-VM project, which brings LIO active-active transport recovery
and other Linux storage functionality to Vmware and Qemu images that can
provide target mode IP storage fabric on x86 non-linux based hosts. A
first of its kind in the Linux/iSCSI universe.
Anyways, lets get back to the technical discussion.
--nab
> Bart Van Assche.
>
David Dillow wrote:
> On Thu, 2008-01-31 at 18:08 +0100, Bart Van Assche wrote:
>
>>If anyone has a suggestion for a better test than dd to compare the
>>performance of SCSI storage protocols, please let it know.
>
>
> xdd on /dev/sda, sdb, etc. using -dio to do direct IO seems to work
> decently, though it is hard (ie, impossible) to get a repeatable
> sequence of IO when using higher queue depths, as it uses threads to
> generate multiple requests.
This utility seems to be a good one, but it's basically the same as
disktest, although much more advanced.
> You may also look at sgpdd_survey from Lustre's iokit, but I've not done
> much with that -- it uses the sg devices to send lowlevel SCSI commands.
Yes, it might be worth to try. Since fundamentally it's the same as
O_DIRECT dd, but with a bit less overhead on the initiator side (hence
less initiator side latency), most likely it will show ever bigger
difference, than it is with dd.
> I've been playing around with some benchmark code using libaio, but it's
> not in generally usable shape.
>
> xdd:
> http://www.ioperformance.com/products.htm
>
> Lustre IO Kit:
> http://manual.lustre.org/manual/LustreManual16_HTML/DynamicHTML-20-1.html
Vladislav Bolkhovitin wrote:
> Bart Van Assche wrote:
>
>> On Jan 31, 2008 5:25 PM, Joe Landman <[email protected]>
>> wrote:
>>
>>> Vladislav Bolkhovitin wrote:
>>>
>>>> Actually, I don't know what kind of conclusions it is possible to make
>>>> from disktest's results (maybe only how throughput gets bigger or
>>>> slower
>>>> with increasing number of threads?), it's a good stress test tool, but
>>>> not more.
>>>
>>>
>>> Unfortunately, I agree. Bonnie++, dd tests, and a few others seem to
>>> bear far closer to "real world" tests than disktest and iozone, the
>>> latter of which does more to test the speed of RAM cache and system call
>>> performance than actual IO.
>>
>>
>>
>> I have ran some tests with Bonnie++, but found out that on a fast
>> network like IB the filesystem used for the test has a really big
>> impact on the test results.
>>
>> If anyone has a suggestion for a better test than dd to compare the
>> performance of SCSI storage protocols, please let it know.
>
>
> I would suggest you to try something from real life, like:
>
> - Copying large file tree over a single or multiple IB links
>
> - Measure of some DB engine's TPC
>
> - etc.
Forgot to mention. During those tests make sure that imported devices
from both SCST and STGT report in the kernel log the same write cache
and FUA capabilities, since they significantly affect initiator's
behavior. Like:
sd 4:0:0:5: [sdf] Write cache: enabled, read cache: enabled, supports
DPO and FUA
For SCST the fastest mode is NV_CACHE, refer to its README file for details.
>> Bart Van Assche.
>>
>> -------------------------------------------------------------------------
>> This SF.net email is sponsored by: Microsoft
>> Defy all challenges. Microsoft(R) Visual Studio 2008.
>> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
>> _______________________________________________
>> Scst-devel mailing list
>> [email protected]
>> https://lists.sourceforge.net/lists/listinfo/scst-devel
>>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
On Feb 1, 2008 1:05 PM, Nicholas A. Bellinger <[email protected]> wrote:
>
> All of the kernel and C userspace code is open source and available from
> linux-iscsi.org and licensed under the GPL.
I found a statement on a web page that the ERL2 implementation is not
included in the GPL version (http://zaal.org/iscsi/index.html). The
above implies that this statement is incorrect. Tomo, are you the
maintainer of this web page ?
I'll try to measure the performance of the LIO Target Stack on the
same setup on which I ran the other performance tests.
Bart Van Assche.
On Fri, 2008-02-01 at 14:25 +0100, Bart Van Assche wrote:
> On Feb 1, 2008 1:05 PM, Nicholas A. Bellinger <[email protected]> wrote:
> >
> > All of the kernel and C userspace code is open source and available from
> > linux-iscsi.org and licensed under the GPL.
>
> I found a statement on a web page that the ERL2 implementation is not
> included in the GPL version (http://zaal.org/iscsi/index.html). The
> above implies that this statement is incorrect. Tomo, are you the
> maintainer of this web page ?
>
This was mentioned in the context of the Core-iSCSI Initiator module.
> I'll try to measure the performance of the LIO Target Stack on the
> same setup on which I ran the other performance tests.
>
Great!
--nab
[email protected] wrote on Wed, 30 Jan 2008 10:34 -0600:
> On Wed, 2008-01-30 at 09:38 +0100, Bart Van Assche wrote:
> > On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
> > >
> > > iSER has parameters to limit the maximum size of RDMA (it needs to
> > > repeat RDMA with a poor configuration)?
> >
> > Please specify which parameters you are referring to. As you know I
> > had already repeated my tests with ridiculously high values for the
> > following iSER parameters: FirstBurstLength, MaxBurstLength and
> > MaxRecvDataSegmentLength (16 MB, which is more than the 1 MB block
> > size specified to dd).
>
> the 1Mb block size is a bit of a red herring. Unless you've
> specifically increased the max_sector_size and are using an sg_chain
> converted driver, on x86 the maximum possible transfer accumulation is
> 0.5MB.
>
> I certainly don't rule out that increasing the transfer size up from
> 0.5MB might be the way to improve STGT efficiency, since at an 1GB/s
> theoretical peak, that's roughly 2000 context switches per I/O; however,
> It doesn't look like you've done anything that will overcome the block
> layer limitations.
The MRDSL parameter has no effect on iSER, as the RFC describes.
How to transfer data to satisfy a command is solely up to the
target. So you would need both big requests from the client, then
look at how the target will send the data.
I've only used 512 kB for the RDMA transfer size from the target, as
it matches the default client size and was enough to get good
performance out of my IB gear and minimizes resource consumption on
the target. It's currently hard-coded as a #define. There is no
provision in the protocol for the client to dictate the value.
If others want to spend some effort trying to tune stgt for iSER,
there are a fair number of comments in the code, including a big one
that explains this RDMA transfer size issue. And I'll answer
informed questions as I can. But I'm not particularly interested in
arguing about which implementation is best, or trying to interpret
bandwidth comparison numbers from poorly designed tests. It takes
work to understand these issues.
-- Pete
Vladislav Bolkhovitin wrote:
> James Bottomley wrote:
>
>> The two target architectures perform essentially identical functions, so
>> there's only really room for one in the kernel. Right at the moment,
>> it's STGT. Problems in STGT come from the user<->kernel boundary which
>> can be mitigated in a variety of ways. The fact that the figures are
>> pretty much comparable on non IB networks shows this.
>>
>> I really need a whole lot more evidence than at worst a 20% performance
>> difference on IB to pull one implementation out and replace it with
>> another. Particularly as there's no real evidence that STGT can't be
>> tweaked to recover the 20% even on IB.
>
>
> James,
>
> Although the performance difference between STGT and SCST is apparent,
> this isn't the only point why SCST is better. I've already written about
> it many times in various mailing lists, but let me summarize it one more
> time here.
>
> As you know, almost all kernel parts can be done in user space,
> including all the drivers, networking, I/O management with block/SCSI
> initiator subsystem and disk cache manager. But does it mean that
> currently Linux kernel is bad and all the above should be (re)done in
> user space instead? I believe, not. Linux isn't a microkernel for very
> pragmatic reasons: simplicity and performance. So, additional important
> point why SCST is better is simplicity.
>
> For SCSI target, especially with hardware target card, data are came
> from kernel and eventually served by kernel, which does actual I/O or
> getting/putting data from/to cache. Dividing requests processing between
> user and kernel space creates unnecessary interface layer(s) and
> effectively makes the requests processing job distributed with all its
> complexity and reliability problems. From my point of view, having such
> distribution, where user space is master side and kernel is slave is
> rather wrong, because:
>
> 1. It makes kernel depend from user program, which services it and
> provides for it its routines, while the regular paradigm is the
> opposite: kernel services user space applications. As a direct
> consequence from it that there is no real protection for the kernel from
> faults in the STGT core code without excessive effort, which, no
> surprise, wasn't currently done and, seems, is never going to be done.
> So, on practice debugging and developing under STGT isn't easier, than
> if the whole code was in the kernel space, but, actually, harder (see
> below why).
>
> 2. It requires new complicated interface between kernel and user spaces
> that creates additional maintenance and debugging headaches, which don't
> exist for kernel only code. Linus Torvalds some time ago perfectly
> described why it is bad, see http://lkml.org/lkml/2007/4/24/451,
> http://lkml.org/lkml/2006/7/1/41 and http://lkml.org/lkml/2007/4/24/364.
>
> 3. It makes for SCSI target impossible to use (at least, on a simple and
> sane way) many effective optimizations: zero-copy cached I/O, more
> control over read-ahead, device queue unplugging-plugging, etc. One
> example of already implemented such features is zero-copy network data
> transmission, done in simple 260 lines put_page_callback patch. This
> optimization is especially important for the user space gate (scst_user
> module), see below for details.
>
> The whole point that development for kernel is harder, than for user
> space, is totally nonsense nowadays. It's different, yes, in some ways
> more limited, yes, but not harder. For ones who need gdb (I for many
> years - don't) kernel has kgdb, plus it also has many not available for
> user space or more limited there debug facilities like lockdep, lockup
> detection, oprofile, etc. (I don't mention wider choice of more
> effectively implemented synchronization primitives and not only them).
>
> For people who need complicated target devices emulation, like, e.g., in
> case of VTL (Virtual Tape Library), where there is a need to operate
> with large mmap'ed memory areas, SCST provides gateway to the user space
> (scst_user module), but, in contrast with STGT, it's done in regular
> "kernel - master, user application - slave" paradigm, so it's reliable
> and no fault in user space device emulator can break kernel and other
> user space applications. Plus, since SCSI target state machine and
> memory management are in the kernel, it's very effective and allows only
> one kernel-user space switch per SCSI command.
>
> Also, I should note here, that in the current state STGT in many aspects
> doesn't fully conform SCSI specifications, especially in area of
> management events, like Unit Attentions generation and processing, and
> it doesn't look like somebody cares about it. At the same time, SCST
> pays big attention to fully conform SCSI specifications, because price
> of non-conformance is a possible user's data corruption.
>
> Returning to performance, modern SCSI transports, e.g. InfiniBand, have
> as low link latency as 1(!) microsecond. For comparison, the
> inter-thread context switch time on a modern system is about the same,
> syscall time - about 0.1 microsecond. So, only ten empty syscalls or one
> context switch add the same latency as the link. Even 1Gbps Ethernet has
> less, than 100 microseconds of round-trip latency.
>
> You, probably, know, that QLogic Fibre Channel target driver for SCST
> allows commands being executed either directly from soft IRQ, or from
> the corresponding thread. There is a steady 5-7% difference in IOPS
> between those modes on 512 bytes reads on nullio using 4Gbps link. So, a
> single additional inter-kernel-thread context switch costs 5-7% of IOPS.
>
> Another source of additional unavoidable with the user space approach
> latency is data copy to/from cache. With the fully kernel space
> approach, cache can be used directly, so no extra copy will be needed.
> We can estimate how much latency the data copying adds. On the modern
> systems memory copy throughput is less than 2GB/s, so on 20Gbps
> InfiniBand link it almost doubles data transfer latency.
>
> So, putting code in the user space you should accept the extra latency
> it adds. Many, if not most, real-life workloads more or less latency,
> not throughput, bound, so there shouldn't be surprise that single stream
> "dd if=/dev/sdX of=/dev/null" on initiator gives too low values. Such
> "benchmark" isn't less important and practical, than all the
> multithreaded latency insensitive benchmarks, which people like running,
> because it does essentially the same as most Linux processes do when
> they read data from files.
>
> You may object me that the target's backstorage device(s) latency is a
> lot more, than 1 microsecond, but that is relevant only if data are
> read/written from/to the actual backstorage media, not from the cache,
> even from the backstorage device's cache. Nothing prevents target from
> having 8 or even 64GB of cache, so most even random accesses could be
> served by it. This is especially important for sync writes.
>
> Thus, why SCST is better:
>
> 1. It is more simple, because it's monolithic, so all its components are
> in one place and communicate using direct function calls. Hence, it is
> smaller, faster, more reliable and maintainable. Currently it's bigger,
> than STGT, just because it supports more features, see (2).
>
> 2. It supports more features: 1 to many pass-through support with all
> necessary for it functionality, including support for non-disk SCSI
> devices, like tapes, SGV cache, BLOCKIO, where requests converted to
> bio's and directly sent to block level (this mode is effective for
> random mostly workloads with data set size >> memory size on the
> target), etc.
>
> 3. It has better performance and going to have it even better. SCST only
> now enters in the phase, where it starts exploiting all advantages of
> being in the kernel. Particularly, zero-copy cached I/O is currently
> being implemented.
>
> 4. It provides safer and more effective interface to emulate target
> devices in the user space via scst_user module.
>
> 5. It much more confirms to SCSI specifications (see above).
So, James, what is your opinion on the above? Or the overall SCSI target
project simplicity doesn't matter much for you and you think it's fine
to duplicate Linux page cache in the user space to keep the in-kernel
part of the project as small as possible?
Vlad
On Feb 4, 2008 1:27 PM, Vladislav Bolkhovitin <[email protected]> wrote:
>
> So, James, what is your opinion on the above? Or the overall SCSI target
> project simplicity doesn't matter much for you and you think it's fine
> to duplicate Linux page cache in the user space to keep the in-kernel
> part of the project as small as possible?
It's too early to draw conclusions about performance. I'm currently
performing more measurements, and the results are not easy to
interpret. My plan is to measure the following:
* Setup: target with RAM disk of 2 GB as backing storage.
* Throughput reported by dd and xdd (direct I/O).
* Transfers with dd/xdd in units of 1 KB to 1 GB (the smallest
transfer size that can be specified to xdd is 1 KB).
* Target SCSI software to be tested: IETD iSCSI via IPoIB, STGT iSCSI
via IPoIB, STGT iSER, SCST iSCSI via IPoIB, SCST SRP, LIO iSCSI via
IPoIB.
The reason I chose dd/xdd for these tests is that I want to measure
the performance of the communication protocols, and that I am assuming
that this performance can be modeled by the following formula:
(transfer time in s) = (transfer setup latency in s) + (transfer size
in MB) / (bandwidth in MB/s). Measuring the time needed for transfers
with varying block size allows to compute the constants in the above
formula via linear regression.
One difficulty I already encountered is that the performance of the
Linux IPoIB implementation varies a lot under high load
(http://bugzilla.kernel.org/show_bug.cgi?id=9883).
Another issue I have to look further into is that dd and xdd report
different results for very large block sizes (> 1 MB).
Bart Van Assche.
On Mon, 2008-02-04 at 15:27 +0300, Vladislav Bolkhovitin wrote:
> Vladislav Bolkhovitin wrote:
> So, James, what is your opinion on the above? Or the overall SCSI target
> project simplicity doesn't matter much for you and you think it's fine
> to duplicate Linux page cache in the user space to keep the in-kernel
> part of the project as small as possible?
The answers were pretty much contained here
http://marc.info/?l=linux-scsi&m=120164008302435
and here:
http://marc.info/?l=linux-scsi&m=120171067107293
Weren't they?
James
James Bottomley wrote:
>>Vladislav Bolkhovitin wrote:
>>So, James, what is your opinion on the above? Or the overall SCSI target
>>project simplicity doesn't matter much for you and you think it's fine
>>to duplicate Linux page cache in the user space to keep the in-kernel
>>part of the project as small as possible?
>
>
> The answers were pretty much contained here
>
> http://marc.info/?l=linux-scsi&m=120164008302435
>
> and here:
>
> http://marc.info/?l=linux-scsi&m=120171067107293
>
> Weren't they?
No, sorry, it doesn't look so for me. They are about performance, but
I'm asking about the overall project's architecture, namely about one
part of it: simplicity. Particularly, what do you think about
duplicating Linux page cache in the user space to have zero-copy cached
I/O? Or can you suggest another architectural solution for that problem
in the STGT's approach?
Vlad
On Mon, 2008-02-04 at 14:53 +0100, Bart Van Assche wrote:
> Another issue I have to look further into is that dd and xdd report
> different results for very large block sizes (> 1 MB).
Be aware that xdd reports 1 MB as 1000000, not 1048576. Though, it looks
like dd is the same, so that's probably not helpful. Also, make sure
you're passing {i,o}flag=direct to dd if you're using -dio in xdd to be
sure you are comparing apples to apples.
--
Dave Dillow
National Center for Computational Science
Oak Ridge National Laboratory
(865) 241-6602 office
On Mon, 2008-02-04 at 19:25 +0300, Vladislav Bolkhovitin wrote:
> James Bottomley wrote:
> >>Vladislav Bolkhovitin wrote:
> >>So, James, what is your opinion on the above? Or the overall SCSI target
> >>project simplicity doesn't matter much for you and you think it's fine
> >>to duplicate Linux page cache in the user space to keep the in-kernel
> >>part of the project as small as possible?
> >
> >
> > The answers were pretty much contained here
> >
> > http://marc.info/?l=linux-scsi&m=120164008302435
> >
> > and here:
> >
> > http://marc.info/?l=linux-scsi&m=120171067107293
> >
> > Weren't they?
>
> No, sorry, it doesn't look so for me. They are about performance, but
> I'm asking about the overall project's architecture, namely about one
> part of it: simplicity. Particularly, what do you think about
> duplicating Linux page cache in the user space to have zero-copy cached
> I/O? Or can you suggest another architectural solution for that problem
> in the STGT's approach?
Isn't that an advantage of a user space solution? It simply uses the
backing store of whatever device supplies the data. That means it takes
advantage of the existing mechanisms for caching.
James
Bart Van Assche wrote:
> On Feb 4, 2008 1:27 PM, Vladislav Bolkhovitin <[email protected]> wrote:
>
>>So, James, what is your opinion on the above? Or the overall SCSI target
>>project simplicity doesn't matter much for you and you think it's fine
>>to duplicate Linux page cache in the user space to keep the in-kernel
>>part of the project as small as possible?
>
>
> It's too early to draw conclusions about performance. I'm currently
> performing more measurements, and the results are not easy to
> interpret. My plan is to measure the following:
> * Setup: target with RAM disk of 2 GB as backing storage.
> * Throughput reported by dd and xdd (direct I/O).
> * Transfers with dd/xdd in units of 1 KB to 1 GB (the smallest
> transfer size that can be specified to xdd is 1 KB).
> * Target SCSI software to be tested: IETD iSCSI via IPoIB, STGT iSCSI
> via IPoIB, STGT iSER, SCST iSCSI via IPoIB, SCST SRP, LIO iSCSI via
> IPoIB.
>
> The reason I chose dd/xdd for these tests is that I want to measure
> the performance of the communication protocols, and that I am assuming
> that this performance can be modeled by the following formula:
> (transfer time in s) = (transfer setup latency in s) + (transfer size
> in MB) / (bandwidth in MB/s).
It isn't fully correct, you forgot about link latency. More correct one is:
(transfer time) = (transfer setup latency on both initiator and target,
consisting from software processing time, including memory copy, if
necessary, and PCI setup/transfer time) + (transfer size)/(bandwidth) +
(link latency to deliver request for READs or status for WRITES) +
(2*(link latency) to deliver R2T/XFER_READY request in case of WRITEs,
if necessary (e.g. iSER for small transfers might not need it, but SRP
most likely always needs it)). Also you should note that it's correct
only in case of single threaded workloads with one outstanding command
at time. For other workloads it depends from how well they manage to
keep the "link" full in interval from (transfer size)/(transfer time) to
bandwidth.
> Measuring the time needed for transfers
> with varying block size allows to compute the constants in the above
> formula via linear regression.
Unfortunately, it isn't so easy, see above.
> One difficulty I already encountered is that the performance of the
> Linux IPoIB implementation varies a lot under high load
> (http://bugzilla.kernel.org/show_bug.cgi?id=9883).
>
> Another issue I have to look further into is that dd and xdd report
> different results for very large block sizes (> 1 MB).
Look at /proc/scsi_tgt/sgv (for SCST) and you will see, which transfer
sizes are actually used. Initiators don't like sending big requests and
often split them on smaller ones.
Look at this message as well, it might be helpful:
http://lkml.org/lkml/2007/5/16/223
> Bart Van Assche.
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
James Bottomley wrote:
>>>>So, James, what is your opinion on the above? Or the overall SCSI target
>>>>project simplicity doesn't matter much for you and you think it's fine
>>>>to duplicate Linux page cache in the user space to keep the in-kernel
>>>>part of the project as small as possible?
>>>
>>>
>>>The answers were pretty much contained here
>>>
>>>http://marc.info/?l=linux-scsi&m=120164008302435
>>>
>>>and here:
>>>
>>>http://marc.info/?l=linux-scsi&m=120171067107293
>>>
>>>Weren't they?
>>
>>No, sorry, it doesn't look so for me. They are about performance, but
>>I'm asking about the overall project's architecture, namely about one
>>part of it: simplicity. Particularly, what do you think about
>>duplicating Linux page cache in the user space to have zero-copy cached
>>I/O? Or can you suggest another architectural solution for that problem
>>in the STGT's approach?
>
>
> Isn't that an advantage of a user space solution? It simply uses the
> backing store of whatever device supplies the data. That means it takes
> advantage of the existing mechanisms for caching.
No, please reread this thread, especially this message:
http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
the advantages of the kernel space implementation. The user space
implementation has to have data copied between the cache and user space
buffer, but the kernel space one can use pages in the cache directly,
without extra copy.
Vlad
On Mon, 2008-02-04 at 20:16 +0300, Vladislav Bolkhovitin wrote:
> James Bottomley wrote:
> >>>>So, James, what is your opinion on the above? Or the overall SCSI target
> >>>>project simplicity doesn't matter much for you and you think it's fine
> >>>>to duplicate Linux page cache in the user space to keep the in-kernel
> >>>>part of the project as small as possible?
> >>>
> >>>
> >>>The answers were pretty much contained here
> >>>
> >>>http://marc.info/?l=linux-scsi&m=120164008302435
> >>>
> >>>and here:
> >>>
> >>>http://marc.info/?l=linux-scsi&m=120171067107293
> >>>
> >>>Weren't they?
> >>
> >>No, sorry, it doesn't look so for me. They are about performance, but
> >>I'm asking about the overall project's architecture, namely about one
> >>part of it: simplicity. Particularly, what do you think about
> >>duplicating Linux page cache in the user space to have zero-copy cached
> >>I/O? Or can you suggest another architectural solution for that problem
> >>in the STGT's approach?
> >
> >
> > Isn't that an advantage of a user space solution? It simply uses the
> > backing store of whatever device supplies the data. That means it takes
> > advantage of the existing mechanisms for caching.
>
> No, please reread this thread, especially this message:
> http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
> the advantages of the kernel space implementation. The user space
> implementation has to have data copied between the cache and user space
> buffer, but the kernel space one can use pages in the cache directly,
> without extra copy.
Well, you've said it thrice (the bellman cried) but that doesn't make it
true.
The way a user space solution should work is to schedule mmapped I/O
from the backing store and then send this mmapped region off for target
I/O. For reads, the page gather will ensure that the pages are up to
date from the backing store to the cache before sending the I/O out.
For writes, You actually have to do a msync on the region to get the
data secured to the backing store. You also have to pull tricks with
the mmap region in the case of writes to prevent useless data being read
in from the backing store. However, none of this involves data copies.
James
James Bottomley wrote:
> On Mon, 2008-02-04 at 20:16 +0300, Vladislav Bolkhovitin wrote:
>
>>James Bottomley wrote:
>>
>>>>>>So, James, what is your opinion on the above? Or the overall SCSI target
>>>>>>project simplicity doesn't matter much for you and you think it's fine
>>>>>>to duplicate Linux page cache in the user space to keep the in-kernel
>>>>>>part of the project as small as possible?
>>>>>
>>>>>
>>>>>The answers were pretty much contained here
>>>>>
>>>>>http://marc.info/?l=linux-scsi&m=120164008302435
>>>>>
>>>>>and here:
>>>>>
>>>>>http://marc.info/?l=linux-scsi&m=120171067107293
>>>>>
>>>>>Weren't they?
>>>>
>>>>No, sorry, it doesn't look so for me. They are about performance, but
>>>>I'm asking about the overall project's architecture, namely about one
>>>>part of it: simplicity. Particularly, what do you think about
>>>>duplicating Linux page cache in the user space to have zero-copy cached
>>>>I/O? Or can you suggest another architectural solution for that problem
>>>>in the STGT's approach?
>>>
>>>
>>>Isn't that an advantage of a user space solution? It simply uses the
>>>backing store of whatever device supplies the data. That means it takes
>>>advantage of the existing mechanisms for caching.
>>
>>No, please reread this thread, especially this message:
>>http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
>>the advantages of the kernel space implementation. The user space
>>implementation has to have data copied between the cache and user space
>>buffer, but the kernel space one can use pages in the cache directly,
>>without extra copy.
>
>
> Well, you've said it thrice (the bellman cried) but that doesn't make it
> true.
>
> The way a user space solution should work is to schedule mmapped I/O
> from the backing store and then send this mmapped region off for target
> I/O. For reads, the page gather will ensure that the pages are up to
> date from the backing store to the cache before sending the I/O out.
> For writes, You actually have to do a msync on the region to get the
> data secured to the backing store.
James, have you checked how fast is mmaped I/O if work size > size of
RAM? It's several times slower comparing to buffered I/O. It was many
times discussed in LKML and, seems, VM people consider it unavoidable.
So, using mmaped IO isn't an option for high performance. Plus, mmaped
IO isn't an option for high reliability requirements, since it doesn't
provide a practical way to handle I/O errors.
> You also have to pull tricks with
> the mmap region in the case of writes to prevent useless data being read
> in from the backing store.
Can you be more exact and specify what kind of tricks should be done for
that?
> However, none of this involves data copies.
>
> James
>
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
On Mon, 2008-02-04 at 20:56 +0300, Vladislav Bolkhovitin wrote:
> James Bottomley wrote:
> > On Mon, 2008-02-04 at 20:16 +0300, Vladislav Bolkhovitin wrote:
> >
> >>James Bottomley wrote:
> >>
> >>>>>>So, James, what is your opinion on the above? Or the overall SCSI target
> >>>>>>project simplicity doesn't matter much for you and you think it's fine
> >>>>>>to duplicate Linux page cache in the user space to keep the in-kernel
> >>>>>>part of the project as small as possible?
> >>>>>
> >>>>>
> >>>>>The answers were pretty much contained here
> >>>>>
> >>>>>http://marc.info/?l=linux-scsi&m=120164008302435
> >>>>>
> >>>>>and here:
> >>>>>
> >>>>>http://marc.info/?l=linux-scsi&m=120171067107293
> >>>>>
> >>>>>Weren't they?
> >>>>
> >>>>No, sorry, it doesn't look so for me. They are about performance, but
> >>>>I'm asking about the overall project's architecture, namely about one
> >>>>part of it: simplicity. Particularly, what do you think about
> >>>>duplicating Linux page cache in the user space to have zero-copy cached
> >>>>I/O? Or can you suggest another architectural solution for that problem
> >>>>in the STGT's approach?
> >>>
> >>>
> >>>Isn't that an advantage of a user space solution? It simply uses the
> >>>backing store of whatever device supplies the data. That means it takes
> >>>advantage of the existing mechanisms for caching.
> >>
> >>No, please reread this thread, especially this message:
> >>http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
> >>the advantages of the kernel space implementation. The user space
> >>implementation has to have data copied between the cache and user space
> >>buffer, but the kernel space one can use pages in the cache directly,
> >>without extra copy.
> >
> >
> > Well, you've said it thrice (the bellman cried) but that doesn't make it
> > true.
> >
> > The way a user space solution should work is to schedule mmapped I/O
> > from the backing store and then send this mmapped region off for target
> > I/O. For reads, the page gather will ensure that the pages are up to
> > date from the backing store to the cache before sending the I/O out.
> > For writes, You actually have to do a msync on the region to get the
> > data secured to the backing store.
>
> James, have you checked how fast is mmaped I/O if work size > size of
> RAM? It's several times slower comparing to buffered I/O. It was many
> times discussed in LKML and, seems, VM people consider it unavoidable.
Erm, but if you're using the case of work size > size of RAM, you'll
find buffered I/O won't help because you don't have the memory for
buffers either.
> So, using mmaped IO isn't an option for high performance. Plus, mmaped
> IO isn't an option for high reliability requirements, since it doesn't
> provide a practical way to handle I/O errors.
I think you'll find it does ... the page gather returns -EFAULT if
there's an I/O error in the gathered region. msync does something
similar if there's a write failure.
> > You also have to pull tricks with
> > the mmap region in the case of writes to prevent useless data being read
> > in from the backing store.
>
> Can you be more exact and specify what kind of tricks should be done for
> that?
Actually, just avoid touching it seems to do the trick with a recent
kernel.
James
On Mon, 4 Feb 2008, James Bottomley wrote:
>
> The way a user space solution should work is to schedule mmapped I/O
> from the backing store and then send this mmapped region off for target
> I/O.
mmap'ing may avoid the copy, but the overhead of a mmap operation is
quite often much *bigger* than the overhead of a copy operation.
Please do not advocate the use of mmap() as a way to avoid memory copies.
It's not realistic. Even if you can do it with a single "mmap()" system
call (which is not at all a given, considering that block devices can
easily be much larger than the available virtual memory space), the fact
is that page table games along with the fault (and even just TLB miss)
overhead is easily more than the cost of copying a page in a nice
streaming manner.
Yes, memory is "slow", but dammit, so is mmap().
> You also have to pull tricks with the mmap region in the case of writes
> to prevent useless data being read in from the backing store. However,
> none of this involves data copies.
"data copies" is irrelevant. The only thing that matters is performance.
And if avoiding data copies is more costly (or even of a similar cost)
than the copies themselves would have been, there is absolutely no upside,
and only downsides due to extra complexity.
If you want good performance for a service like this, you really generally
*do* need to in kernel space. You can play games in user space, but you're
fooling yourself if you think you can do as well as doing it in the
kernel. And you're *definitely* fooling yourself if you think mmap()
solves performance issues. "Zero-copy" does not equate to "fast". Memory
speeds may be slower that core CPU speeds, but not infinitely so!
(That said: there *are* alternatives to mmap, like "splice()", that really
do potentially solve some issues without the page table and TLB overheads.
But while splice() avoids the costs of paging, I strongly suspect it would
still have easily measurable latency issues. Switching between user and
kernel space multiple times is definitely not going to be free, although
it's probably not a huge issue if you have big enough requests).
Linus
James Bottomley wrote:
> On Mon, 2008-02-04 at 20:56 +0300, Vladislav Bolkhovitin wrote:
>
>>James Bottomley wrote:
>>
>>>On Mon, 2008-02-04 at 20:16 +0300, Vladislav Bolkhovitin wrote:
>>>
>>>
>>>>James Bottomley wrote:
>>>>
>>>>
>>>>>>>>So, James, what is your opinion on the above? Or the overall SCSI target
>>>>>>>>project simplicity doesn't matter much for you and you think it's fine
>>>>>>>>to duplicate Linux page cache in the user space to keep the in-kernel
>>>>>>>>part of the project as small as possible?
>>>>>>>
>>>>>>>
>>>>>>>The answers were pretty much contained here
>>>>>>>
>>>>>>>http://marc.info/?l=linux-scsi&m=120164008302435
>>>>>>>
>>>>>>>and here:
>>>>>>>
>>>>>>>http://marc.info/?l=linux-scsi&m=120171067107293
>>>>>>>
>>>>>>>Weren't they?
>>>>>>
>>>>>>No, sorry, it doesn't look so for me. They are about performance, but
>>>>>>I'm asking about the overall project's architecture, namely about one
>>>>>>part of it: simplicity. Particularly, what do you think about
>>>>>>duplicating Linux page cache in the user space to have zero-copy cached
>>>>>>I/O? Or can you suggest another architectural solution for that problem
>>>>>>in the STGT's approach?
>>>>>
>>>>>
>>>>>Isn't that an advantage of a user space solution? It simply uses the
>>>>>backing store of whatever device supplies the data. That means it takes
>>>>>advantage of the existing mechanisms for caching.
>>>>
>>>>No, please reread this thread, especially this message:
>>>>http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
>>>>the advantages of the kernel space implementation. The user space
>>>>implementation has to have data copied between the cache and user space
>>>>buffer, but the kernel space one can use pages in the cache directly,
>>>>without extra copy.
>>>
>>>
>>>Well, you've said it thrice (the bellman cried) but that doesn't make it
>>>true.
>>>
>>>The way a user space solution should work is to schedule mmapped I/O
>>>from the backing store and then send this mmapped region off for target
>>>I/O. For reads, the page gather will ensure that the pages are up to
>>>date from the backing store to the cache before sending the I/O out.
>>>For writes, You actually have to do a msync on the region to get the
>>>data secured to the backing store.
>>
>>James, have you checked how fast is mmaped I/O if work size > size of
>>RAM? It's several times slower comparing to buffered I/O. It was many
>>times discussed in LKML and, seems, VM people consider it unavoidable.
>
>
> Erm, but if you're using the case of work size > size of RAM, you'll
> find buffered I/O won't help because you don't have the memory for
> buffers either.
James, just check and you will see, buffered I/O is a lot faster.
>>So, using mmaped IO isn't an option for high performance. Plus, mmaped
>>IO isn't an option for high reliability requirements, since it doesn't
>>provide a practical way to handle I/O errors.
>
> I think you'll find it does ... the page gather returns -EFAULT if
> there's an I/O error in the gathered region.
Err, to whom return? If you try to read from a mmaped page, which can't
be populated due to I/O error, you will get SIGBUS or SIGSEGV, I don't
remember exactly. It's quite tricky to get back to the faulted command
from the signal handler.
Or do you mean mmap(MAP_POPULATE)/munmap() for each command? Do you
think that such mapping/unmapping is good for performance?
> msync does something
> similar if there's a write failure.
>
>>>You also have to pull tricks with
>>>the mmap region in the case of writes to prevent useless data being read
>>>in from the backing store.
>>
>>Can you be more exact and specify what kind of tricks should be done for
>>that?
>
> Actually, just avoid touching it seems to do the trick with a recent
> kernel.
Hmm, how can one write to an mmaped page and don't touch it?
> James
>
>
>
On Mon, 2008-02-04 at 10:29 -0800, Linus Torvalds wrote:
>
> On Mon, 4 Feb 2008, James Bottomley wrote:
> >
> > The way a user space solution should work is to schedule mmapped I/O
> > from the backing store and then send this mmapped region off for target
> > I/O.
>
> mmap'ing may avoid the copy, but the overhead of a mmap operation is
> quite often much *bigger* than the overhead of a copy operation.
>
> Please do not advocate the use of mmap() as a way to avoid memory copies.
> It's not realistic. Even if you can do it with a single "mmap()" system
> call (which is not at all a given, considering that block devices can
> easily be much larger than the available virtual memory space), the fact
> is that page table games along with the fault (and even just TLB miss)
> overhead is easily more than the cost of copying a page in a nice
> streaming manner.
>
> Yes, memory is "slow", but dammit, so is mmap().
>
> > You also have to pull tricks with the mmap region in the case of writes
> > to prevent useless data being read in from the backing store. However,
> > none of this involves data copies.
>
> "data copies" is irrelevant. The only thing that matters is performance.
> And if avoiding data copies is more costly (or even of a similar cost)
> than the copies themselves would have been, there is absolutely no upside,
> and only downsides due to extra complexity.
>
> If you want good performance for a service like this, you really generally
> *do* need to in kernel space. You can play games in user space, but you're
> fooling yourself if you think you can do as well as doing it in the
> kernel. And you're *definitely* fooling yourself if you think mmap()
> solves performance issues. "Zero-copy" does not equate to "fast". Memory
> speeds may be slower that core CPU speeds, but not infinitely so!
>
> (That said: there *are* alternatives to mmap, like "splice()", that really
> do potentially solve some issues without the page table and TLB overheads.
> But while splice() avoids the costs of paging, I strongly suspect it would
> still have easily measurable latency issues. Switching between user and
> kernel space multiple times is definitely not going to be free, although
> it's probably not a huge issue if you have big enough requests).
Sorry ... this is really just a discussion of how something (zero copy)
could be done, rather than an implementation proposal. (I'm not
actually planning to make the STGT people do anything ... although
investigating splice does sound interesting).
Right at the moment, STGT seems to be performing just fine on
measurements up to gigabit networks. There are suggestions that there
may be a problem on 8G IB networks, but it's not definitive yet.
I'm already on record as saying I think the best fix for IB networks is
just to reduce the context switches by increasing the transfer size, but
the infrastructure to allow that only just went into git head.
James
On Mon, 2008-02-04 at 21:38 +0300, Vladislav Bolkhovitin wrote:
> James Bottomley wrote:
> > On Mon, 2008-02-04 at 20:56 +0300, Vladislav Bolkhovitin wrote:
> >
> >>James Bottomley wrote:
> >>
> >>>On Mon, 2008-02-04 at 20:16 +0300, Vladislav Bolkhovitin wrote:
> >>>
> >>>
> >>>>James Bottomley wrote:
> >>>>
> >>>>
> >>>>>>>>So, James, what is your opinion on the above? Or the overall SCSI target
> >>>>>>>>project simplicity doesn't matter much for you and you think it's fine
> >>>>>>>>to duplicate Linux page cache in the user space to keep the in-kernel
> >>>>>>>>part of the project as small as possible?
> >>>>>>>
> >>>>>>>
> >>>>>>>The answers were pretty much contained here
> >>>>>>>
> >>>>>>>http://marc.info/?l=linux-scsi&m=120164008302435
> >>>>>>>
> >>>>>>>and here:
> >>>>>>>
> >>>>>>>http://marc.info/?l=linux-scsi&m=120171067107293
> >>>>>>>
> >>>>>>>Weren't they?
> >>>>>>
> >>>>>>No, sorry, it doesn't look so for me. They are about performance, but
> >>>>>>I'm asking about the overall project's architecture, namely about one
> >>>>>>part of it: simplicity. Particularly, what do you think about
> >>>>>>duplicating Linux page cache in the user space to have zero-copy cached
> >>>>>>I/O? Or can you suggest another architectural solution for that problem
> >>>>>>in the STGT's approach?
> >>>>>
> >>>>>
> >>>>>Isn't that an advantage of a user space solution? It simply uses the
> >>>>>backing store of whatever device supplies the data. That means it takes
> >>>>>advantage of the existing mechanisms for caching.
> >>>>
> >>>>No, please reread this thread, especially this message:
> >>>>http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
> >>>>the advantages of the kernel space implementation. The user space
> >>>>implementation has to have data copied between the cache and user space
> >>>>buffer, but the kernel space one can use pages in the cache directly,
> >>>>without extra copy.
> >>>
> >>>
> >>>Well, you've said it thrice (the bellman cried) but that doesn't make it
> >>>true.
> >>>
> >>>The way a user space solution should work is to schedule mmapped I/O
> >>>from the backing store and then send this mmapped region off for target
> >>>I/O. For reads, the page gather will ensure that the pages are up to
> >>>date from the backing store to the cache before sending the I/O out.
> >>>For writes, You actually have to do a msync on the region to get the
> >>>data secured to the backing store.
> >>
> >>James, have you checked how fast is mmaped I/O if work size > size of
> >>RAM? It's several times slower comparing to buffered I/O. It was many
> >>times discussed in LKML and, seems, VM people consider it unavoidable.
> >
> >
> > Erm, but if you're using the case of work size > size of RAM, you'll
> > find buffered I/O won't help because you don't have the memory for
> > buffers either.
>
> James, just check and you will see, buffered I/O is a lot faster.
So in an out of memory situation the buffers you don't have are a lot
faster than the pages I don't have?
> >>So, using mmaped IO isn't an option for high performance. Plus, mmaped
> >>IO isn't an option for high reliability requirements, since it doesn't
> >>provide a practical way to handle I/O errors.
> >
> > I think you'll find it does ... the page gather returns -EFAULT if
> > there's an I/O error in the gathered region.
>
> Err, to whom return? If you try to read from a mmaped page, which can't
> be populated due to I/O error, you will get SIGBUS or SIGSEGV, I don't
> remember exactly. It's quite tricky to get back to the faulted command
> from the signal handler.
>
> Or do you mean mmap(MAP_POPULATE)/munmap() for each command? Do you
> think that such mapping/unmapping is good for performance?
>
> > msync does something
> > similar if there's a write failure.
> >
> >>>You also have to pull tricks with
> >>>the mmap region in the case of writes to prevent useless data being read
> >>>in from the backing store.
> >>
> >>Can you be more exact and specify what kind of tricks should be done for
> >>that?
> >
> > Actually, just avoid touching it seems to do the trick with a recent
> > kernel.
>
> Hmm, how can one write to an mmaped page and don't touch it?
I meant from user space ... the writes are done inside the kernel.
However, as Linus has pointed out, this discussion is getting a bit off
topic. There's no actual evidence that copy problems are causing any
performatince issues issues for STGT. In fact, there's evidence that
they're not for everything except IB networks.
James
On Mon, 2008-02-04 at 10:29 -0800, Linus Torvalds wrote:
>
> On Mon, 4 Feb 2008, James Bottomley wrote:
> >
> > The way a user space solution should work is to schedule mmapped I/O
> > from the backing store and then send this mmapped region off for target
> > I/O.
>
> mmap'ing may avoid the copy, but the overhead of a mmap operation is
> quite often much *bigger* than the overhead of a copy operation.
>
> Please do not advocate the use of mmap() as a way to avoid memory copies.
> It's not realistic. Even if you can do it with a single "mmap()" system
> call (which is not at all a given, considering that block devices can
> easily be much larger than the available virtual memory space), the fact
> is that page table games along with the fault (and even just TLB miss)
> overhead is easily more than the cost of copying a page in a nice
> streaming manner.
>
> Yes, memory is "slow", but dammit, so is mmap().
>
> > You also have to pull tricks with the mmap region in the case of writes
> > to prevent useless data being read in from the backing store. However,
> > none of this involves data copies.
>
> "data copies" is irrelevant. The only thing that matters is performance.
> And if avoiding data copies is more costly (or even of a similar cost)
> than the copies themselves would have been, there is absolutely no upside,
> and only downsides due to extra complexity.
>
The iSER spec (RFC-5046) quotes the following in the TCP case for direct
data placement:
" Out-of-order TCP segments in the Traditional iSCSI model have to be
stored and reassembled before the iSCSI protocol layer within an end
node can place the data in the iSCSI buffers. This reassembly is
required because not every TCP segment is likely to contain an iSCSI
header to enable its placement, and TCP itself does not have a
built-in mechanism for signaling Upper Level Protocol (ULP) message
boundaries to aid placement of out-of-order segments. This TCP
reassembly at high network speeds is quite counter-productive for the
following reasons: wasted memory bandwidth in data copying, the need
for reassembly memory, wasted CPU cycles in data copying, and the
general store-and-forward latency from an application perspective."
While this does not have anything to do directly with the kernel vs. user discussion
for target mode storage engine, the scaling and latency case is easy enough
to make if we are talking about scaling TCP for 10 Gb/sec storage fabrics.
> If you want good performance for a service like this, you really generally
> *do* need to in kernel space. You can play games in user space, but you're
> fooling yourself if you think you can do as well as doing it in the
> kernel. And you're *definitely* fooling yourself if you think mmap()
> solves performance issues. "Zero-copy" does not equate to "fast". Memory
> speeds may be slower that core CPU speeds, but not infinitely so!
>
>From looking at this problem from a kernel space perspective for a
number of years, I would be inclined to believe this is true for
software and hardware data-path cases. The benefits of moving various
control statemachines for something like say traditional iSCSI to
userspace has always been debateable. The most obvious ones are things
like authentication, espically if something more complex than CHAP are
the obvious case for userspace. However, I have thought recovery for
failures caused from communication path (iSCSI connections) or entire
nexuses (iSCSI sessions) failures was very problematic to expect to have
to potentially push down IOs state to userspace.
Keeping statemachines for protocol and/or fabric specific statemachines
(CSM-E and CSM-I from connection recovery in iSCSI and iSER are the
obvious ones) are the best canidates for residing in kernel space.
> (That said: there *are* alternatives to mmap, like "splice()", that really
> do potentially solve some issues without the page table and TLB overheads.
> But while splice() avoids the costs of paging, I strongly suspect it would
> still have easily measurable latency issues. Switching between user and
> kernel space multiple times is definitely not going to be free, although
> it's probably not a huge issue if you have big enough requests).
>
Most of the SCSI OS storage subsystems that I have worked with in the
context of iSCSI have used 256 * 512 byte setctor requests, which the
default traditional iSCSI PDU data payload (MRDSL) being 64k to hit the
sweet spot with crc32c checksum calculations. I am assuming this is
going to be the case for other fabrics as well.
--nab
On Mon, 2008-02-04 at 11:06 -0800, Nicholas A. Bellinger wrote:
> On Mon, 2008-02-04 at 10:29 -0800, Linus Torvalds wrote:
> >
> > On Mon, 4 Feb 2008, James Bottomley wrote:
> > >
> > > The way a user space solution should work is to schedule mmapped I/O
> > > from the backing store and then send this mmapped region off for target
> > > I/O.
> >
> > mmap'ing may avoid the copy, but the overhead of a mmap operation is
> > quite often much *bigger* than the overhead of a copy operation.
> >
> > Please do not advocate the use of mmap() as a way to avoid memory copies.
> > It's not realistic. Even if you can do it with a single "mmap()" system
> > call (which is not at all a given, considering that block devices can
> > easily be much larger than the available virtual memory space), the fact
> > is that page table games along with the fault (and even just TLB miss)
> > overhead is easily more than the cost of copying a page in a nice
> > streaming manner.
> >
> > Yes, memory is "slow", but dammit, so is mmap().
> >
> > > You also have to pull tricks with the mmap region in the case of writes
> > > to prevent useless data being read in from the backing store. However,
> > > none of this involves data copies.
> >
> > "data copies" is irrelevant. The only thing that matters is performance.
> > And if avoiding data copies is more costly (or even of a similar cost)
> > than the copies themselves would have been, there is absolutely no upside,
> > and only downsides due to extra complexity.
> >
>
> The iSER spec (RFC-5046) quotes the following in the TCP case for direct
> data placement:
>
> " Out-of-order TCP segments in the Traditional iSCSI model have to be
> stored and reassembled before the iSCSI protocol layer within an end
> node can place the data in the iSCSI buffers. This reassembly is
> required because not every TCP segment is likely to contain an iSCSI
> header to enable its placement, and TCP itself does not have a
> built-in mechanism for signaling Upper Level Protocol (ULP) message
> boundaries to aid placement of out-of-order segments. This TCP
> reassembly at high network speeds is quite counter-productive for the
> following reasons: wasted memory bandwidth in data copying, the need
> for reassembly memory, wasted CPU cycles in data copying, and the
> general store-and-forward latency from an application perspective."
>
> While this does not have anything to do directly with the kernel vs. user discussion
> for target mode storage engine, the scaling and latency case is easy enough
> to make if we are talking about scaling TCP for 10 Gb/sec storage fabrics.
>
> > If you want good performance for a service like this, you really generally
> > *do* need to in kernel space. You can play games in user space, but you're
> > fooling yourself if you think you can do as well as doing it in the
> > kernel. And you're *definitely* fooling yourself if you think mmap()
> > solves performance issues. "Zero-copy" does not equate to "fast". Memory
> > speeds may be slower that core CPU speeds, but not infinitely so!
> >
>
> >From looking at this problem from a kernel space perspective for a
> number of years, I would be inclined to believe this is true for
> software and hardware data-path cases. The benefits of moving various
> control statemachines for something like say traditional iSCSI to
> userspace has always been debateable. The most obvious ones are things
> like authentication, espically if something more complex than CHAP are
> the obvious case for userspace. However, I have thought recovery for
> failures caused from communication path (iSCSI connections) or entire
> nexuses (iSCSI sessions) failures was very problematic to expect to have
> to potentially push down IOs state to userspace.
>
> Keeping statemachines for protocol and/or fabric specific statemachines
> (CSM-E and CSM-I from connection recovery in iSCSI and iSER are the
> obvious ones) are the best canidates for residing in kernel space.
>
> > (That said: there *are* alternatives to mmap, like "splice()", that really
> > do potentially solve some issues without the page table and TLB overheads.
> > But while splice() avoids the costs of paging, I strongly suspect it would
> > still have easily measurable latency issues. Switching between user and
> > kernel space multiple times is definitely not going to be free, although
> > it's probably not a huge issue if you have big enough requests).
> >
>
Then again, having some data-path for software and hardware bulk IO
operation of storage fabric protocol / statemachine in userspace would
be really interesting for something like an SPU enabled engine for the
Cell Broadband Architecture.
--nab
On Mon, 4 Feb 2008, Nicholas A. Bellinger wrote:
>
> While this does not have anything to do directly with the kernel vs.
> user discussion for target mode storage engine, the scaling and latency
> case is easy enough to make if we are talking about scaling TCP for 10
> Gb/sec storage fabrics.
I would like to point out that while I think there is no question that the
basic data transfer engine would perform better in kernel space, there
stll *are* questions whether
- iSCSI is relevant enough for us to even care ...
- ... and the complexity is actually worth it.
That said, I also tend to believe that trying to split things up between
kernel and user space is often more complex than just keeping things in
one place, because the trade-offs of which part goes where wll inevitably
be wrong in *some* area, and then you're really screwed.
So from a purely personal standpoint, I'd like to say that I'm not really
interested in iSCSI (and I don't quite know why I've been cc'd on this
whole discussion) and think that other approaches are potentially *much*
better. So for example, I personally suspect that ATA-over-ethernet is way
better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
low-level, and against those crazy SCSI people to begin with.
So take any utterances of mine with a big pinch of salt.
Historically, the only split that has worked pretty well is "connection
initiation/setup in user space, actual data transfers in kernel space".
Pure user-space solutions work, but tend to eventually be turned into
kernel-space if they are simple enough and really do have throughput and
latency considerations (eg nfsd), and aren't quite complex and crazy
enough to have a large impedance-matching problem even for basic IO stuff
(eg samba).
And totally pure kernel solutions work only if there are very stable
standards and no major authentication or connection setup issues (eg local
disks).
So just going by what has happened in the past, I'd assume that iSCSI
would eventually turn into "connecting/authentication in user space" with
"data transfers in kernel space". But only if it really does end up
mattering enough. We had a totally user-space NFS daemon for a long time,
and it was perfectly fine until people really started caring.
Linus
On luned? 4 febbraio 2008, Linus Torvalds wrote:
> So from a purely personal standpoint, I'd like to say that I'm not really
> interested in iSCSI (and I don't quite know why I've been cc'd on this
> whole discussion) and think that other approaches are potentially *much*
> better. So for example, I personally suspect that ATA-over-ethernet is way
> better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> low-level, and against those crazy SCSI people to begin with.
surely aoe is better than iscsi almost on performance because of the lesser
protocol stack:
iscsi -> scsi - ip - eth
aoe -> ata - eth
but surely iscsi is more a standard than aoe and is more actively used by
real-world .
Other really useful feature are that:
- iscsi is capable to move to a ip based san scsi devices by routing that (
i've some tape changer routed by scst to some system that don't have other
way to see a tape).
- because it work on the ip layer it can be routed between long distance , so
having needed bandwidth you can have a really remote block device spoking a
standard protocol between non ethereogenus systems.
- iscsi is now the cheapest san avaible.
bye,
marco.
On Mon, 2008-02-04 at 11:44 -0800, Linus Torvalds wrote:
>
> On Mon, 4 Feb 2008, Nicholas A. Bellinger wrote:
> >
> > While this does not have anything to do directly with the kernel vs.
> > user discussion for target mode storage engine, the scaling and latency
> > case is easy enough to make if we are talking about scaling TCP for 10
> > Gb/sec storage fabrics.
>
> I would like to point out that while I think there is no question that the
> basic data transfer engine would perform better in kernel space, there
> stll *are* questions whether
>
> - iSCSI is relevant enough for us to even care ...
>
> - ... and the complexity is actually worth it.
>
> That said, I also tend to believe that trying to split things up between
> kernel and user space is often more complex than just keeping things in
> one place, because the trade-offs of which part goes where wll inevitably
> be wrong in *some* area, and then you're really screwed.
>
> So from a purely personal standpoint, I'd like to say that I'm not really
> interested in iSCSI (and I don't quite know why I've been cc'd on this
> whole discussion)
The generic target mode storage engine discussion quickly goes to
transport specific scenarios. With so much interest in the SCSI
transports, in particuarly iSCSI, there are lots of devs, users, and
vendors who would like to see Linux improve in this respect.
> and think that other approaches are potentially *much*
> better. So for example, I personally suspect that ATA-over-ethernet is way
> better than some crazy SCSI-over-TCP crap,
Having the non SCSI target mode transports use the same data IO path as
the SCSI ones to SCSI, BIO, and FILE subsystems is something that can
easily be agreed on. Also having to emulate the non SCSI control paths
in a non generic matter to a target mode engine has to suck (I don't
know what AoE does for that now, considering that this is going down to
libata or real SCSI hardware in some cases. There are some of the more
arcane task management functionality in SCSI (ACA anyone?) that even
generic SCSI target mode engines do not use, and only seem to make
endlessly complex implement and emulate.
But aside from those very SCSI hardware specific cases, having a generic
method to use something like ABORT_TASK or LUN_RESET for a target mode
engine (along with the data path to all of the subsystems) would be
beneficial for any fabric.
> but I'm biased for simple and
> low-level, and against those crazy SCSI people to begin with.
Well, having no obvious preconception (well, aside from the email
address), I am of the mindset than the iSCSI people are the LEAST crazy
said crazy SCSI people. Some people (usually least crazy iSCSI
standards folks) say that FCoE people are crazy. Being one of the iSCSI
people I am kinda obligated to agree, but the technical points are
really solid, and have been so for over a decade. They are listed here
for those who are interested:
http://www.ietf.org/mail-archive/web/ips/current/msg02325.html
>
> So take any utterances of mine with a big pinch of salt.
>
> Historically, the only split that has worked pretty well is "connection
> initiation/setup in user space, actual data transfers in kernel space".
>
> Pure user-space solutions work, but tend to eventually be turned into
> kernel-space if they are simple enough and really do have throughput and
> latency considerations (eg nfsd), and aren't quite complex and crazy
> enough to have a large impedance-matching problem even for basic IO stuff
> (eg samba).
>
> And totally pure kernel solutions work only if there are very stable
> standards and no major authentication or connection setup issues (eg local
> disks).
>
> So just going by what has happened in the past, I'd assume that iSCSI
> would eventually turn into "connecting/authentication in user space" with
> "data transfers in kernel space". But only if it really does end up
> mattering enough. We had a totally user-space NFS daemon for a long time,
> and it was perfectly fine until people really started caring.
Thanks for putting this into an historical perspective. Also it is
interesting to note that the iSCSI spec (RFC-3720) was ratified in April
2004, so it will be going on 4 years soon, which pre-RFC products first
going out in 2001 (yikes!). In my experience, the iSCSI interopt
amongst implementations (espically between different OSes) has been
stable since about late 2004, early 2005, with interopt between OS SCSI
subsystems (espically talking to non SCSI hardware) being the slower of
the two.
--nab
On Mon, Feb 04, 2008 at 11:44:31AM -0800, Linus Torvalds wrote:
...
> Pure user-space solutions work, but tend to eventually be turned into
> kernel-space if they are simple enough and really do have throughput and
> latency considerations (eg nfsd), and aren't quite complex and crazy
> enough to have a large impedance-matching problem even for basic IO stuff
> (eg samba).
...
> So just going by what has happened in the past, I'd assume that iSCSI
> would eventually turn into "connecting/authentication in user space" with
> "data transfers in kernel space". But only if it really does end up
> mattering enough. We had a totally user-space NFS daemon for a long time,
> and it was perfectly fine until people really started caring.
I'd assumed the move was primarily because of the difficulty of getting
correct semantics on a shared filesystem--if you're content with
NFS-only access to your filesystem, then you can probably do everything
in userspace, but once you start worrying about getting stable
filehandles, consistent file locking, etc., from a real disk filesystem
with local users, then you require much closer cooperation from the
kernel.
And I seem to recall being told that sort of thing was the motivation
more than performance, but I wasn't there (and I haven't seen
performance comparisons).
--b.
On Mon, 4 Feb 2008, J. Bruce Fields wrote:
>
> I'd assumed the move was primarily because of the difficulty of getting
> correct semantics on a shared filesystem
.. not even shared. It was hard to get correct semantics full stop.
Which is a traditional problem. The thing is, the kernel always has some
internal state, and it's hard to expose all the semantics that the kernel
knows about to user space.
So no, performance is not the only reason to move to kernel space. It can
easily be things like needing direct access to internal data queues (for a
iSCSI target, this could be things like barriers or just tagged commands -
yes, you can probably emulate things like that without access to the
actual IO queues, but are you sure the semantics will be entirely right?
The kernel/userland boundary is not just a performance boundary, it's an
abstraction boundary too, and these kinds of protocols tend to break
abstractions. NFS broke it by having "file handles" (which is not
something that really exists in user space, and is almost impossible to
emulate correctly), and I bet the same thing happens when emulating a SCSI
target in user space.
Maybe not. I _rally_ haven't looked into iSCSI, I'm just guessing there
would be things like ordering issues.
Linus
On Mon, 2008-02-04 at 13:24 -0800, Linus Torvalds wrote:
>
> On Mon, 4 Feb 2008, J. Bruce Fields wrote:
> >
> > I'd assumed the move was primarily because of the difficulty of getting
> > correct semantics on a shared filesystem
>
> .. not even shared. It was hard to get correct semantics full stop.
>
> Which is a traditional problem. The thing is, the kernel always has some
> internal state, and it's hard to expose all the semantics that the kernel
> knows about to user space.
>
> So no, performance is not the only reason to move to kernel space. It can
> easily be things like needing direct access to internal data queues (for a
> iSCSI target, this could be things like barriers or just tagged commands -
> yes, you can probably emulate things like that without access to the
> actual IO queues, but are you sure the semantics will be entirely right?
>
> The kernel/userland boundary is not just a performance boundary, it's an
> abstraction boundary too, and these kinds of protocols tend to break
> abstractions. NFS broke it by having "file handles" (which is not
> something that really exists in user space, and is almost impossible to
> emulate correctly), and I bet the same thing happens when emulating a SCSI
> target in user space.
>
> Maybe not. I _rally_ haven't looked into iSCSI, I'm just guessing there
> would be things like ordering issues.
>
<nod>.
The iSCSI CDBs and write immediate, unsoliciated, or soliciated data
payloads may be received out of order across communication paths (which
may be going over different subnets) within the nexus, but the execution
of the CDB to SCSI Target Port must be in the same order as it came down
from the SCSI subsystem on the initiator port. In iSCSI and iSER terms,
this is called Command Sequence Number (CmdSN) ordering, and is enforced
within each nexus. The initiator node will be assigning the CmdSNs as
the CDBs come down, and when communication paths fail, unacknowledged
CmdSNs will be retried on a different communication path when using
iSCSI/iSER connection recovery. Already acknowledged CmdSNs will be
explictly retried using a iSCSI specific task management function called
TASK_REASSIGN. This along with CSM-I and CSM-E statemachines are
collectly known as ErrorRecoveryLevel=2 in iSCSI.
Anyways, here is a great visual of a modern iSCSI Target processor and
SCSI Target Engine. The CmdSN ordering is representd by the oval across
across iSCSI connections going to various network portals groups on the
left side of the diagram. Thanks Eddy Q!
http://www.haifa.il.ibm.com/satran/ips/EddyQuicksall-iSCSI-in-diagrams/portal_groups.pdf
--nab
> better. So for example, I personally suspect that ATA-over-ethernet is way
> better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> low-level, and against those crazy SCSI people to begin with.
Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
would probably trash iSCSI for latency if nothing else.
Alan
Linus Torvalds wrote:
> So no, performance is not the only reason to move to kernel space. It can
> easily be things like needing direct access to internal data queues (for a
> iSCSI target, this could be things like barriers or just tagged commands -
> yes, you can probably emulate things like that without access to the
> actual IO queues, but are you sure the semantics will be entirely right?
>
> The kernel/userland boundary is not just a performance boundary, it's an
> abstraction boundary too, and these kinds of protocols tend to break
> abstractions. NFS broke it by having "file handles" (which is not
> something that really exists in user space, and is almost impossible to
> emulate correctly), and I bet the same thing happens when emulating a SCSI
> target in user space.
Well, speaking as a complete nutter who just finished the bare bones of
an NFSv4 userland server[1]... it depends on your approach.
If the userland server is the _only_ one accessing the data[2] -- i.e.
the database server model where ls(1) shows a couple multi-gigabyte
files or a raw partition -- then it's easy to get all the semantics
right, including file handles. You're not racing with local kernel
fileserving.
Couple that with sendfile(2), sync_file_range(2) and a few other
Linux-specific syscalls, and you've got an efficient NFS file server.
It becomes a solution similar to Apache or MySQL or Oracle.
I quite grant there are many good reasons to do NFS or iSCSI data path
in the kernel... my point is more that "impossible" is just from one
point of view ;-)
> Maybe not. I _rally_ haven't looked into iSCSI, I'm just guessing there
> would be things like ordering issues.
iSCSI and NBD were passe ideas at birth. :)
Networked block devices are attractive because the concepts and
implementation are more simple than networked filesystems... but usually
you want to run some sort of filesystem on top. At that point you might
as well run NFS or [gfs|ocfs|flavor-of-the-week], and ditch your
networked block device (and associated complexity).
iSCSI is barely useful, because at least someone finally standardized
SCSI over LAN/WAN.
But you just don't need its complexity if your filesystem must have its
own authentication, distributed coordination, multiple-connection
management code of its own.
Jeff
P.S. Clearly my NFSv4 server is NOT intended to replace the kernel one.
It's more for experiments, and doing FUSE-like filesystem work.
[1] http://linux.yyz.us/projects/nfsv4.html
[2] well, outside of dd(1) and similar tricks... the same "going around
its back" tricks that can screw up a mounted filesystem.
On Mon, 2008-02-04 at 22:43 +0000, Alan Cox wrote:
> > better. So for example, I personally suspect that ATA-over-ethernet is way
> > better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > low-level, and against those crazy SCSI people to begin with.
>
> Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> would probably trash iSCSI for latency if nothing else.
>
In the previous iSCSI vs. FCoE points (here is the link again):
http://www.ietf.org/mail-archive/web/ips/current/msg02325.html
the latency discussion is the one bit that is not mentioned. I always
assumed that back then (as with today) the biggest issue was getting
ethernet hardware, espically switching equipment down to the sub
millisecond latency, and on par with what you would expect from 'real
RDMA' hardware. In lowest of the low, say sub 10 ns latency, which is
apparently possible with point to point on high-end 10 Gb/sec adapters
today, it would be really interesting to know how much more latency
would be expected between software iSCSI vs. *oE when we work our way
back up the networking stack.
Julo, do you have any idea on this..?
--nab
>
> Alan
>
On Mon, 2008-02-04 at 22:43 +0000, Alan Cox wrote:
> > better. So for example, I personally suspect that ATA-over-ethernet is way
> > better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > low-level, and against those crazy SCSI people to begin with.
>
> Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> would probably trash iSCSI for latency if nothing else.
Actually, there's also FCoE now ... which is essentially SCSI
encapsulated in Fibre Channel Protocols (FCP) running over ethernet with
Jumbo frames. It does the standard SCSI TCQ, so should answer all the
latency pieces. Intel even has an implementation:
http://www.open-fcoe.org/
I tend to prefer the low levels as well. The whole disadvantage for IP
as regards iSCSI was the layers of protocols on top of it for
addressing, authenticating, encrypting and finding any iSCSI device
anywhere in the connected universe.
I tend to see loss of routing from operating at the MAC level to be a
nicely justifiable tradeoff (most storage networks tend to be hubbed or
switched anyway). Plus an ethernet MAC with jumbo frames is a large
framed nearly lossless medium, which is practically what FCP is
expecting. If you really have to connect large remote sites ... well
that's what tunnelling bridges are for.
James
Alan Cox wrote:
>> better. So for example, I personally suspect that ATA-over-ethernet is way
>> better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
>> low-level, and against those crazy SCSI people to begin with.
>
> Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> would probably trash iSCSI for latency if nothing else.
AoE is truly a thing of beauty. It has a two/three page RFC (say no more!).
But quite so... AoE is limited to MTU size, which really hurts. Can't
really do tagged queueing, etc.
iSCSI is way, way too complicated. It's an Internet protocol designed
by storage designers, what do you expect?
For years I have been hoping that someone will invent a simple protocol
(w/ strong auth) that can transit ATA and SCSI commands and responses.
Heck, it would be almost trivial if the kernel had a TLS/SSL implementation.
Jeff
On Mon, 2008-02-04 at 17:00 -0600, James Bottomley wrote:
> On Mon, 2008-02-04 at 22:43 +0000, Alan Cox wrote:
> > > better. So for example, I personally suspect that ATA-over-ethernet is way
> > > better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > > low-level, and against those crazy SCSI people to begin with.
> >
> > Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> > would probably trash iSCSI for latency if nothing else.
>
> Actually, there's also FCoE now ... which is essentially SCSI
> encapsulated in Fibre Channel Protocols (FCP) running over ethernet with
> Jumbo frames. It does the standard SCSI TCQ, so should answer all the
> latency pieces. Intel even has an implementation:
>
> http://www.open-fcoe.org/
>
> I tend to prefer the low levels as well. The whole disadvantage for IP
> as regards iSCSI was the layers of protocols on top of it for
> addressing, authenticating, encrypting and finding any iSCSI device
> anywhere in the connected universe.
Btw, while simple in-band discovery of iSCSI exists, the standards based
IP storage deployments (iSCSI and iFCP) use iSNS (RFC-4171) for
discovery and network fabric management, for things like sending state
change notifications when a particular network portal is going away so
that the initiator can bring up a different communication patch to a
different network portal, etc.
>
> I tend to see loss of routing from operating at the MAC level to be a
> nicely justifiable tradeoff (most storage networks tend to be hubbed or
> switched anyway). Plus an ethernet MAC with jumbo frames is a large
> framed nearly lossless medium, which is practically what FCP is
> expecting. If you really have to connect large remote sites ... well
> that's what tunnelling bridges are for.
>
Some of the points by Julo on the IPS TWG iSCSI vs. FCoE thread:
* the network is limited in physical span and logical span (number
of switches)
* flow-control/congestion control is achieved with a mechanism
adequate for a limited span network (credits). The packet loss
rate is almost nil and that allows FCP to avoid using a
transport (end-to-end) layer
* FCP she switches are simple (addresses are local and the memory
requirements cam be limited through the credit mechanism)
* The credit mechanisms is highly unstable for large networks
(check switch vendors planning docs for the network diameter
limits) – the scaling argument
* Ethernet has no credit mechanism and any mechanism with a
similar effect increases the end point cost. Building a
transport layer in the protocol stack has always been the
preferred choice of the networking community – the community
argument
* The "performance penalty" of a complete protocol stack has
always been overstated (and overrated). Advances in protocol
stack implementation and finer tuning of the congestion control
mechanisms make conventional TCP/IP performing well even at 10
Gb/s and over. Moreover the multicore processors that become
dominant on the computing scene have enough compute cycles
available to make any "offloading" possible as a mere code
restructuring exercise (see the stack reports from Intel, IBM
etc.)
* Building on a complete stack makes available a wealth of
operational and management mechanisms built over the years by
the networking community (routing, provisioning, security,
service location etc.) – the community argument
* Higher level storage access over an IP network is widely
available and having both block and file served over the same
connection with the same support and management structure is
compelling– the community argument
* Highly efficient networks are easy to build over IP with optimal
(shortest path) routing while Layer 2 networks use bridging and
are limited by the logical tree structure that bridges must
follow. The effort to combine routers and bridges (rbridges) is
promising to change that but it will take some time to finalize
(and we don't know exactly how it will operate). Untill then the
scale of Layer 2 network is going to seriously limited – the
scaling argument
Perhaps it would be of worth to get some more linux-net guys in on the
discussion. :-)
--nab
> James
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
On Mon, 2008-02-04 at 15:12 -0800, Nicholas A. Bellinger wrote:
> On Mon, 2008-02-04 at 17:00 -0600, James Bottomley wrote:
> > On Mon, 2008-02-04 at 22:43 +0000, Alan Cox wrote:
> > > > better. So for example, I personally suspect that ATA-over-ethernet is way
> > > > better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > > > low-level, and against those crazy SCSI people to begin with.
> > >
> > > Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> > > would probably trash iSCSI for latency if nothing else.
> >
> > Actually, there's also FCoE now ... which is essentially SCSI
> > encapsulated in Fibre Channel Protocols (FCP) running over ethernet with
> > Jumbo frames. It does the standard SCSI TCQ, so should answer all the
> > latency pieces. Intel even has an implementation:
> >
> > http://www.open-fcoe.org/
> >
> > I tend to prefer the low levels as well. The whole disadvantage for IP
> > as regards iSCSI was the layers of protocols on top of it for
> > addressing, authenticating, encrypting and finding any iSCSI device
> > anywhere in the connected universe.
>
> Btw, while simple in-band discovery of iSCSI exists, the standards based
> IP storage deployments (iSCSI and iFCP) use iSNS (RFC-4171) for
> discovery and network fabric management, for things like sending state
> change notifications when a particular network portal is going away so
> that the initiator can bring up a different communication patch to a
> different network portal, etc.
>
> >
> > I tend to see loss of routing from operating at the MAC level to be a
> > nicely justifiable tradeoff (most storage networks tend to be hubbed or
> > switched anyway). Plus an ethernet MAC with jumbo frames is a large
> > framed nearly lossless medium, which is practically what FCP is
> > expecting. If you really have to connect large remote sites ... well
> > that's what tunnelling bridges are for.
> >
>
> Some of the points by Julo on the IPS TWG iSCSI vs. FCoE thread:
>
> * the network is limited in physical span and logical span (number
> of switches)
> * flow-control/congestion control is achieved with a mechanism
> adequate for a limited span network (credits). The packet loss
> rate is almost nil and that allows FCP to avoid using a
> transport (end-to-end) layer
> * FCP she switches are simple (addresses are local and the memory
> requirements cam be limited through the credit mechanism)
> * The credit mechanisms is highly unstable for large networks
> (check switch vendors planning docs for the network diameter
> limits) – the scaling argument
> * Ethernet has no credit mechanism and any mechanism with a
> similar effect increases the end point cost. Building a
> transport layer in the protocol stack has always been the
> preferred choice of the networking community – the community
> argument
> * The "performance penalty" of a complete protocol stack has
> always been overstated (and overrated). Advances in protocol
> stack implementation and finer tuning of the congestion control
> mechanisms make conventional TCP/IP performing well even at 10
> Gb/s and over. Moreover the multicore processors that become
> dominant on the computing scene have enough compute cycles
> available to make any "offloading" possible as a mere code
> restructuring exercise (see the stack reports from Intel, IBM
> etc.)
> * Building on a complete stack makes available a wealth of
> operational and management mechanisms built over the years by
> the networking community (routing, provisioning, security,
> service location etc.) – the community argument
> * Higher level storage access over an IP network is widely
> available and having both block and file served over the same
> connection with the same support and management structure is
> compelling– the community argument
> * Highly efficient networks are easy to build over IP with optimal
> (shortest path) routing while Layer 2 networks use bridging and
> are limited by the logical tree structure that bridges must
> follow. The effort to combine routers and bridges (rbridges) is
> promising to change that but it will take some time to finalize
> (and we don't know exactly how it will operate). Untill then the
> scale of Layer 2 network is going to seriously limited – the
> scaling argument
>
Another data point from the "The "performance penalty of a complete
protocol stack has always been overstated (and overrated)" bullet above:
"As a side argument – a performance comparison made in 1998 showed SCSI
over TCP (a predecessor of the later iSCSI) to perform better than FCP
at 1Gbs for block sizes typical for OLTP (4-8KB). That was what
convinced us to take the path that lead to iSCSI – and we used plain
vanilla x86 servers with plain-vanilla NICs and Linux (with similar
measurements conducted on Windows)."
--nab
On Mon, 4 Feb 2008, Jeff Garzik wrote:
>
> For years I have been hoping that someone will invent a simple protocol (w/
> strong auth) that can transit ATA and SCSI commands and responses. Heck, it
> would be almost trivial if the kernel had a TLS/SSL implementation.
Why would you want authorization? If you don't use IP (just ethernet
framing), then 99% of the time the solution is to just trust the subnet.
So most people would never want TLS/SSL, and the ones that *do* want it
would probably also want IP routing, so you'd actually be better off with
a separate higher-level bridging protocol rather than have TLS/SSL as part
of the actual packet protocol.
So don't add complexity. The beauty of ATA-over-ethernet is exactly that
it's simple and straightforward.
(Simple and straightforward is also nice for actually creating devices
that are the targets of this. I just *bet* that an iSCSI target device
probably needs two orders of magnitude more CPU power than a simple AoE
thing that can probably be done in an FPGA with no real software at all).
Whatever. We have now officially gotten totally off topic ;)
Linus
Alan Cox wrote:
>> better. So for example, I personally suspect that ATA-over-ethernet is way
>> better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
>> low-level, and against those crazy SCSI people to begin with.
>
> Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> would probably trash iSCSI for latency if nothing else.
And a variant that doesn't do ATA or IP:
http://www.fcoe.com/
On Mon, 4 Feb 2008, Jeff Garzik wrote:
>
> Well, speaking as a complete nutter who just finished the bare bones of an
> NFSv4 userland server[1]... it depends on your approach.
You definitely are a complete nutter ;)
> If the userland server is the _only_ one accessing the data[2] -- i.e. the
> database server model where ls(1) shows a couple multi-gigabyte files or a raw
> partition -- then it's easy to get all the semantics right, including file
> handles. You're not racing with local kernel fileserving.
It's not really simple in general even then. The problems come with file
handles, and two big issues in particular:
- handling a reboot (of the server) without impacting the client really
does need a "look up by file handle" operation (which you can do by
logging the pathname to filehandle translation, but it certainly gets
problematic).
- non-Unix-like filesystems don't necessarily have a stable "st_ino"
field (ie it may change over a rename or have no meaning what-so-ever,
things like that), and that makes trying to generate a filehandle
really interesting for them.
I do agree that it's possible - we obviously _did_ have a user-level NFSD
for a long while, after all - but it's quite painful if you want to handle
things well. Only allowing access through the NFSD certainly helps a lot,
but still doesn't make it quite as trivial as you claim ;)
Of course, I think you can make NFSv4 to use volatile filehandles instead
of the traditional long-lived ones, and that really should avoid almost
all of the problems with doing a NFSv4 server in user space. However, I'd
expect there to be clients that don't do the whole volatile thing, or
support the file handle becoming stale only at certain well-defined points
(ie after renames, not at random reboot times).
Linus
Linus Torvalds wrote:
>
> On Mon, 4 Feb 2008, Jeff Garzik wrote:
>> Well, speaking as a complete nutter who just finished the bare bones of an
>> NFSv4 userland server[1]... it depends on your approach.
>
> You definitely are a complete nutter ;)
>
>> If the userland server is the _only_ one accessing the data[2] -- i.e. the
>> database server model where ls(1) shows a couple multi-gigabyte files or a raw
>> partition -- then it's easy to get all the semantics right, including file
>> handles. You're not racing with local kernel fileserving.
>
> It's not really simple in general even then. The problems come with file
> handles, and two big issues in particular:
>
> - handling a reboot (of the server) without impacting the client really
> does need a "look up by file handle" operation (which you can do by
> logging the pathname to filehandle translation, but it certainly gets
> problematic).
>
> - non-Unix-like filesystems don't necessarily have a stable "st_ino"
> field (ie it may change over a rename or have no meaning what-so-ever,
> things like that), and that makes trying to generate a filehandle
> really interesting for them.
Both of these are easily handled if the server is 100% in charge of
managing the filesystem _metadata_ and data. That's what I meant by
complete control.
i.e. it not ext3 or reiserfs or vfat, its a block device or 1000GB file
managed by a userland process.
Doing it that way gives one a bit more freedom to tune the filesystem
format directly. Stable inode numbers and filehandles are just easy as
they are with ext3. I'm the filesystem format designer, after all. (run
for your lives...)
You do wind up having to roll your own dcache in userspace, though.
A matter of taste in implementation, but it is not difficult... I've
certainly never been accused of having good taste :)
> I do agree that it's possible - we obviously _did_ have a user-level NFSD
> for a long while, after all - but it's quite painful if you want to handle
> things well. Only allowing access through the NFSD certainly helps a lot,
> but still doesn't make it quite as trivial as you claim ;)
Nah, you're thinking about something different: a userland NFSD
competing with other userland processes for access to the same files,
while the kernel ultimately manages the filesystem metadata. Recipe for
races and inequities, and it's good we moved away from that.
I'm talking about where a userland process manages the filesystem
metadata too. In a filesystem with a million files, ls(1) on the server
will only show a single file:
[jgarzik@core ~]$ ls -l /spare/fileserver-data/
total 70657116
-rw-r--r-- 1 jgarzik jgarzik 1818064825 2007-12-29 06:40 fsimage.1
> Of course, I think you can make NFSv4 to use volatile filehandles instead
> of the traditional long-lived ones, and that really should avoid almost
> all of the problems with doing a NFSv4 server in user space. However, I'd
> expect there to be clients that don't do the whole volatile thing, or
> support the file handle becoming stale only at certain well-defined points
> (ie after renames, not at random reboot times).
Don't get me started on "volatile" versus "persistent" filehandles in
NFSv4... groan.
Jeff
On Mon, 2008-02-04 at 22:43 +0000, Alan Cox wrote:
> > better. So for example, I personally suspect that ATA-over-ethernet is way
> > better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > low-level, and against those crazy SCSI people to begin with.
>
> Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> would probably trash iSCSI for latency if nothing else.
But ATAoE is boring because it's not IP. Which means no routing,
firewalls, tunnels, congestion control, etc.
NBD and iSCSI (for all its hideous growths) can take advantage of these
things.
--
Mathematics is the supreme nostalgia of our time.
On Mon, 4 Feb 2008, Matt Mackall wrote:
>
> But ATAoE is boring because it's not IP. Which means no routing,
> firewalls, tunnels, congestion control, etc.
The thing is, that's often an advantage. Not just for performance.
> NBD and iSCSI (for all its hideous growths) can take advantage of these
> things.
.. and all this could equally well be done by a simple bridging protocol
(completely independently of any AoE code).
The thing is, iSCSI does things at the wrong level. It *forces* people to
use the complex protocols, when it's a known that a lot of people don't
want it.
Which is why these AoE and FCoE things keep popping up.
It's easy to bridge ethernet and add a new layer on top of AoE if you need
it. In comparison, it's *impossible* to remove an unnecessary layer from
iSCSI.
This is why "simple and low-level is good". It's always possible to build
on top of low-level protocols, while it's generally never possible to
simplify overly complex ones.
Linus
Linus Torvalds wrote:
> On Mon, 4 Feb 2008, Matt Mackall wrote:
>> But ATAoE is boring because it's not IP. Which means no routing,
>> firewalls, tunnels, congestion control, etc.
>
> The thing is, that's often an advantage. Not just for performance.
>
>> NBD and iSCSI (for all its hideous growths) can take advantage of these
>> things.
>
> .. and all this could equally well be done by a simple bridging protocol
> (completely independently of any AoE code).
>
> The thing is, iSCSI does things at the wrong level. It *forces* people to
> use the complex protocols, when it's a known that a lot of people don't
> want it.
>
> Which is why these AoE and FCoE things keep popping up.
>
> It's easy to bridge ethernet and add a new layer on top of AoE if you need
> it. In comparison, it's *impossible* to remove an unnecessary layer from
> iSCSI.
>
> This is why "simple and low-level is good". It's always possible to build
> on top of low-level protocols, while it's generally never possible to
> simplify overly complex ones.
Never discount "easy" and "just works", which is what IP (and TCP) gives
you...
Sure you can use a bridging protocol and all that jazz, but I wager, to
a network admin yet-another-IP-application is easier to evaluate, deploy
and manage on existing networks.
Jeff
On Mon, 2008-02-04 at 16:24 -0800, Linus Torvalds wrote:
>
> On Mon, 4 Feb 2008, Matt Mackall wrote:
> >
> > But ATAoE is boring because it's not IP. Which means no routing,
> > firewalls, tunnels, congestion control, etc.
>
> The thing is, that's often an advantage. Not just for performance.
>
> > NBD and iSCSI (for all its hideous growths) can take advantage of these
> > things.
>
> .. and all this could equally well be done by a simple bridging protocol
> (completely independently of any AoE code).
>
> The thing is, iSCSI does things at the wrong level. It *forces* people to
> use the complex protocols, when it's a known that a lot of people don't
> want it.
I frankly think NBD is at a pretty comfortable level. It's internally
very simple (and hardware-agnostic). And moderately easy to do in
silicon.
But I'm not going to defend iSCSI. I worked on the first implementation
(what became the Cisco iSCSI driver) and I have no love for iSCSI at
all. It should have been (and started out as) a nearly trivial
encapsulation of SCSI over TCP much like ATA over Ethernet but quickly
lost the plot when committees got ahold of it.
--
Mathematics is the supreme nostalgia of our time.
On Mon, 4 Feb 2008, Jeff Garzik wrote:
>
> Both of these are easily handled if the server is 100% in charge of managing
> the filesystem _metadata_ and data. That's what I meant by complete control.
>
> i.e. it not ext3 or reiserfs or vfat, its a block device or 1000GB file
> managed by a userland process.
Oh ok.
Yes, if you bring the filesystem into user mode too, then the problems go
away - because now your NFSD can interact directly with the filesystem
without any kernel/usermode abstraction layer rules in between. So that
has all the same properties as moving NFSD entirely into the kernel.
Linus
On Feb 4, 2008 11:30 AM, Douglas Gilbert <[email protected]> wrote:
> Alan Cox wrote:
> >> better. So for example, I personally suspect that ATA-over-ethernet is way
> >> better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> >> low-level, and against those crazy SCSI people to begin with.
> >
> > Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> > would probably trash iSCSI for latency if nothing else.
>
> And a variant that doesn't do ATA or IP:
> http://www.fcoe.com/
>
however, and interestingly enough, the open-fcoe software target
depends on scst (for now anyway)
Hi all,
And sorry for intrusion, i am not a developer but i work everyday with iscsi
and i found it fantastic.
Altough Aoe, Fcoe and so on could be better, we have to look in real world
implementations what is needed *now*, and if we look at vmware world,
virtual iron, microsoft clustering etc, the answer is iSCSI.
And now, SCST is the best open-source iSCSI target. So, from an end-user
point of view, what are the really problems to not integrate scst in the
mainstream kernel?
Just my two cent,
--
So long and thank for all the fish
--
#Matteo Tescione
#RMnet srl
>
>
> On Mon, 4 Feb 2008, Matt Mackall wrote:
>>
>> But ATAoE is boring because it's not IP. Which means no routing,
>> firewalls, tunnels, congestion control, etc.
>
> The thing is, that's often an advantage. Not just for performance.
>
>> NBD and iSCSI (for all its hideous growths) can take advantage of these
>> things.
>
> .. and all this could equally well be done by a simple bridging protocol
> (completely independently of any AoE code).
>
> The thing is, iSCSI does things at the wrong level. It *forces* people to
> use the complex protocols, when it's a known that a lot of people don't
> want it.
>
> Which is why these AoE and FCoE things keep popping up.
>
> It's easy to bridge ethernet and add a new layer on top of AoE if you need
> it. In comparison, it's *impossible* to remove an unnecessary layer from
> iSCSI.
>
> This is why "simple and low-level is good". It's always possible to build
> on top of low-level protocols, while it's generally never possible to
> simplify overly complex ones.
>
> Linus
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Scst-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scst-devel
>
On Tue, 2008-02-05 at 05:43 +0100, Matteo Tescione wrote:
> Hi all,
> And sorry for intrusion, i am not a developer but i work everyday with iscsi
> and i found it fantastic.
> Altough Aoe, Fcoe and so on could be better, we have to look in real world
> implementations what is needed *now*, and if we look at vmware world,
> virtual iron, microsoft clustering etc, the answer is iSCSI.
> And now, SCST is the best open-source iSCSI target. So, from an end-user
> point of view, what are the really problems to not integrate scst in the
> mainstream kernel?
The fact that your last statement is conjecture. It's definitely untrue
for non-IB networks, and the jury is still out on IB networks.
James
James Bottomley schrieb:
> These are both features being independently worked on, are they not?
> Even if they weren't, the combination of the size of SCST in kernel plus
> the problem of having to find a migration path for the current STGT
> users still looks to me to involve the greater amount of work.
I don't want to be mean, but does anyone actually use STGT in
production? Seriously?
In the latest development version of STGT, it's only possible to stop
the tgtd target daemon using KILL / 9 signal - which also means all
iSCSI initiator connections are corrupted when tgtd target daemon is
started again (kernel upgrade, target daemon upgrade, server reboot etc.).
Imagine you have to reboot all your NFS clients when you reboot your NFS
server. Not only that - your data is probably corrupted, or at least the
filesystem deserves checking...
--
Tomasz Chmielewski
http://wpkg.org
On Feb 4, 2008 11:57 PM, Jeff Garzik <[email protected]> wrote:
> Networked block devices are attractive because the concepts and
> implementation are more simple than networked filesystems... but usually
> you want to run some sort of filesystem on top. At that point you might
> as well run NFS or [gfs|ocfs|flavor-of-the-week], and ditch your
> networked block device (and associated complexity).
Running a filesystem on top of iSCSI results in better performance
than NFS, especially if the NFS client conforms to the NFS standard
(=synchronous writes).
By searching the web search for the keywords NFS, iSCSI and
performance I found the following (6 years old) document:
http://www.technomagesinc.com/papers/ip_paper.html. A quote from the
conclusion:
Our results, generated by running some of industry standard benchmarks,
show that iSCSI significantly outperforms NFS for situations when
performing streaming, database like accesses and small file transactions.
Bart Van Assche.
On Mon, Feb 04, 2008 at 05:57:47PM -0500, Jeff Garzik wrote:
> iSCSI and NBD were passe ideas at birth. :)
>
> Networked block devices are attractive because the concepts and
> implementation are more simple than networked filesystems... but usually
> you want to run some sort of filesystem on top. At that point you might
> as well run NFS or [gfs|ocfs|flavor-of-the-week], and ditch your
> networked block device (and associated complexity).
Call me a sysadmin, but I find easier to plug in and keep in place an
ethernet cable than these parallel scsi cables from hell. Every
server has at least two ethernet ports by default, with rarely any
surprises at the kernel level. Adding ethernet cards is inexpensive,
and you pretty much never hear of compatibility problems between
cards.
So ethernet as a connection medium is really nice compared to scsi.
Too bad iscsi is demented and ATAoE/NBD inexistant. Maybe external
SAS will be nice, but I don't see it getting to the level of
universality of ethernet any time soon. And it won't get the same
amount of user-level compatibility testing in any case.
OG.
On Tue, 05 Feb 2008 08:14:01 +0100
Tomasz Chmielewski <[email protected]> wrote:
> James Bottomley schrieb:
>
> > These are both features being independently worked on, are they not?
> > Even if they weren't, the combination of the size of SCST in kernel plus
> > the problem of having to find a migration path for the current STGT
> > users still looks to me to involve the greater amount of work.
>
> I don't want to be mean, but does anyone actually use STGT in
> production? Seriously?
>
> In the latest development version of STGT, it's only possible to stop
> the tgtd target daemon using KILL / 9 signal - which also means all
> iSCSI initiator connections are corrupted when tgtd target daemon is
> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
I don't know what "iSCSI initiator connections are corrupted"
mean. But if you reboot a server, how can an iSCSI target
implementation keep iSCSI tcp connections?
> Imagine you have to reboot all your NFS clients when you reboot your NFS
> server. Not only that - your data is probably corrupted, or at least the
> filesystem deserves checking...
On Tue, 05 Feb 2008 05:43:10 +0100
Matteo Tescione <[email protected]> wrote:
> Hi all,
> And sorry for intrusion, i am not a developer but i work everyday with iscsi
> and i found it fantastic.
> Altough Aoe, Fcoe and so on could be better, we have to look in real world
> implementations what is needed *now*, and if we look at vmware world,
> virtual iron, microsoft clustering etc, the answer is iSCSI.
> And now, SCST is the best open-source iSCSI target. So, from an end-user
> point of view, what are the really problems to not integrate scst in the
> mainstream kernel?
Currently, the best open-source iSCSI target implemenation in Linux is
Nicholas's LIO, I guess.
On Mon, 4 Feb 2008 20:07:01 -0600
"Chris Weiss" <[email protected]> wrote:
> On Feb 4, 2008 11:30 AM, Douglas Gilbert <[email protected]> wrote:
> > Alan Cox wrote:
> > >> better. So for example, I personally suspect that ATA-over-ethernet is way
> > >> better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > >> low-level, and against those crazy SCSI people to begin with.
> > >
> > > Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> > > would probably trash iSCSI for latency if nothing else.
> >
> > And a variant that doesn't do ATA or IP:
> > http://www.fcoe.com/
> >
>
> however, and interestingly enough, the open-fcoe software target
> depends on scst (for now anyway)
STGT also supports software FCoE target driver though it's still
experimental stuff.
http://www.mail-archive.com/[email protected]/msg12705.html
It works in user space like STGT's iSCSI (and iSER) target driver
(i.e. no kernel/user space interaction).
FUJITA Tomonori schrieb:
> On Tue, 05 Feb 2008 08:14:01 +0100
> Tomasz Chmielewski <[email protected]> wrote:
>
>> James Bottomley schrieb:
>>
>>> These are both features being independently worked on, are they not?
>>> Even if they weren't, the combination of the size of SCST in kernel plus
>>> the problem of having to find a migration path for the current STGT
>>> users still looks to me to involve the greater amount of work.
>> I don't want to be mean, but does anyone actually use STGT in
>> production? Seriously?
>>
>> In the latest development version of STGT, it's only possible to stop
>> the tgtd target daemon using KILL / 9 signal - which also means all
>> iSCSI initiator connections are corrupted when tgtd target daemon is
>> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
>
> I don't know what "iSCSI initiator connections are corrupted"
> mean. But if you reboot a server, how can an iSCSI target
> implementation keep iSCSI tcp connections?
The problem with tgtd is that you can't start it (configured) in an
"atomic" way.
Usually, one will start tgtd and it's configuration in a script (I
replaced some parameters with "..." to make it shorter and more readable):
tgtd
tgtadm --op new ...
tgtadm --lld iscsi --op new ...
However, this won't work - tgtd goes immediately in the background as it
is still starting, and the first tgtadm commands will fail:
# bash -x tgtd-start
+ tgtd
+ tgtadm --op new --mode target ...
tgtadm: can't connect to the tgt daemon, Connection refused
tgtadm: can't send the request to the tgt daemon, Transport endpoint is
not connected
+ tgtadm --lld iscsi --op new --mode account ...
tgtadm: can't connect to the tgt daemon, Connection refused
tgtadm: can't send the request to the tgt daemon, Transport endpoint is
not connected
+ tgtadm --lld iscsi --op bind --mode account --tid 1 ...
tgtadm: can't find the target
+ tgtadm --op new --mode logicalunit --tid 1 --lun 1 ...
tgtadm: can't find the target
+ tgtadm --op bind --mode target --tid 1 -I ALL
tgtadm: can't find the target
+ tgtadm --op new --mode target --tid 2 ...
+ tgtadm --op new --mode logicalunit --tid 2 --lun 1 ...
+ tgtadm --op bind --mode target --tid 2 -I ALL
OK, if tgtd takes longer to start, perhaps it's a good idea to sleep a
second right after tgtd?
tgtd
sleep 1
tgtadm --op new ...
tgtadm --lld iscsi --op new ...
No, it is not a good idea - if tgtd listens on port 3260 *and* is
unconfigured yet, any reconnecting initiator will fail, like below:
end_request: I/O error, dev sdb, sector 7045192
Buffer I/O error on device sdb, logical block 880649
lost page write due to I/O error on sdb
Aborting journal on device sdb.
ext3_abort called.
EXT3-fs error (device sdb): ext3_journal_start_sb: Detected aborted journal
Remounting filesystem read-only
end_request: I/O error, dev sdb, sector 7045880
Buffer I/O error on device sdb, logical block 880735
lost page write due to I/O error on sdb
end_request: I/O error, dev sdb, sector 6728
Buffer I/O error on device sdb, logical block 841
lost page write due to I/O error on sdb
end_request: I/O error, dev sdb, sector 7045192
Buffer I/O error on device sdb, logical block 880649
lost page write due to I/O error on sdb
end_request: I/O error, dev sdb, sector 7045880
Buffer I/O error on device sdb, logical block 880735
lost page write due to I/O error on sdb
__journal_remove_journal_head: freeing b_frozen_data
__journal_remove_journal_head: freeing b_frozen_data
Ouch.
So the only way to start/restart tgtd reliably is to do hacks which are
needed with yet another iSCSI kernel implementation (IET): use iptables.
iptables <block iSCSI traffic>
tgtd
sleep 1
tgtadm --op new ...
tgtadm --lld iscsi --op new ...
iptables <unblock iSCSI traffic>
A bit ugly, isn't it?
Having to tinker with a firewall in order to start a daemon is by no
means a sign of a well-tested and mature project.
That's why I asked how many people use stgt in a production environment
- James was worried about a potential migration path for current users.
--
Tomasz Chmielewski
http://wpkg.org
On Tue, 2008-02-05 at 17:07 +0100, Tomasz Chmielewski wrote:
> FUJITA Tomonori schrieb:
> > On Tue, 05 Feb 2008 08:14:01 +0100
> > Tomasz Chmielewski <[email protected]> wrote:
> >
> >> James Bottomley schrieb:
> >>
> >>> These are both features being independently worked on, are they not?
> >>> Even if they weren't, the combination of the size of SCST in kernel plus
> >>> the problem of having to find a migration path for the current STGT
> >>> users still looks to me to involve the greater amount of work.
> >> I don't want to be mean, but does anyone actually use STGT in
> >> production? Seriously?
> >>
> >> In the latest development version of STGT, it's only possible to stop
> >> the tgtd target daemon using KILL / 9 signal - which also means all
> >> iSCSI initiator connections are corrupted when tgtd target daemon is
> >> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
> >
> > I don't know what "iSCSI initiator connections are corrupted"
> > mean. But if you reboot a server, how can an iSCSI target
> > implementation keep iSCSI tcp connections?
>
> The problem with tgtd is that you can't start it (configured) in an
> "atomic" way.
> Usually, one will start tgtd and it's configuration in a script (I
> replaced some parameters with "..." to make it shorter and more readable):
>
>
> tgtd
> tgtadm --op new ...
> tgtadm --lld iscsi --op new ...
>
>
> However, this won't work - tgtd goes immediately in the background as it
> is still starting, and the first tgtadm commands will fail:
this should be a easy fix. start tgtd, get port setup ready in forked
process, then signal its parent that ready to quit. or set port ready in
parent, fork and pass to daemon.
>
> # bash -x tgtd-start
> + tgtd
> + tgtadm --op new --mode target ...
> tgtadm: can't connect to the tgt daemon, Connection refused
> tgtadm: can't send the request to the tgt daemon, Transport endpoint is
> not connected
> + tgtadm --lld iscsi --op new --mode account ...
> tgtadm: can't connect to the tgt daemon, Connection refused
> tgtadm: can't send the request to the tgt daemon, Transport endpoint is
> not connected
> + tgtadm --lld iscsi --op bind --mode account --tid 1 ...
> tgtadm: can't find the target
> + tgtadm --op new --mode logicalunit --tid 1 --lun 1 ...
> tgtadm: can't find the target
> + tgtadm --op bind --mode target --tid 1 -I ALL
> tgtadm: can't find the target
> + tgtadm --op new --mode target --tid 2 ...
> + tgtadm --op new --mode logicalunit --tid 2 --lun 1 ...
> + tgtadm --op bind --mode target --tid 2 -I ALL
>
>
> OK, if tgtd takes longer to start, perhaps it's a good idea to sleep a
> second right after tgtd?
>
> tgtd
> sleep 1
> tgtadm --op new ...
> tgtadm --lld iscsi --op new ...
>
>
> No, it is not a good idea - if tgtd listens on port 3260 *and* is
> unconfigured yet, any reconnecting initiator will fail, like below:
this is another easy fix. tgtd started with unconfigured status and then
a tgtadm can configure it and turn it into ready status.
those are really minor usability issue. ( i know it is painful for user,
i agree)
the major problem here is to discuss in architectural wise, which one is
better... linux kernel should have one implementation that is good from
foundation...
>
> end_request: I/O error, dev sdb, sector 7045192
> Buffer I/O error on device sdb, logical block 880649
> lost page write due to I/O error on sdb
> Aborting journal on device sdb.
> ext3_abort called.
> EXT3-fs error (device sdb): ext3_journal_start_sb: Detected aborted journal
> Remounting filesystem read-only
> end_request: I/O error, dev sdb, sector 7045880
> Buffer I/O error on device sdb, logical block 880735
> lost page write due to I/O error on sdb
> end_request: I/O error, dev sdb, sector 6728
> Buffer I/O error on device sdb, logical block 841
> lost page write due to I/O error on sdb
> end_request: I/O error, dev sdb, sector 7045192
> Buffer I/O error on device sdb, logical block 880649
> lost page write due to I/O error on sdb
> end_request: I/O error, dev sdb, sector 7045880
> Buffer I/O error on device sdb, logical block 880735
> lost page write due to I/O error on sdb
> __journal_remove_journal_head: freeing b_frozen_data
> __journal_remove_journal_head: freeing b_frozen_data
>
>
> Ouch.
>
> So the only way to start/restart tgtd reliably is to do hacks which are
> needed with yet another iSCSI kernel implementation (IET): use iptables.
>
> iptables <block iSCSI traffic>
> tgtd
> sleep 1
> tgtadm --op new ...
> tgtadm --lld iscsi --op new ...
> iptables <unblock iSCSI traffic>
>
>
> A bit ugly, isn't it?
> Having to tinker with a firewall in order to start a daemon is by no
> means a sign of a well-tested and mature project.
>
> That's why I asked how many people use stgt in a production environment
> - James was worried about a potential migration path for current users.
>
>
>
> --
> Tomasz Chmielewski
> http://wpkg.org
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Scst-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scst-devel
--
Ming Zhang
@#$%^ purging memory... (*!%
http://blackmagic02881.wordpress.com/
http://www.linkedin.com/in/blackmagic02881
--------------------------------------------
Regarding the performance tests I promised to perform: although until
now I only have been able to run two tests (STGT + iSER versus SCST +
SRP), the results are interesting. I will run the remaining test cases
during the next days.
About the test setup: dd and xdd were used to transfer 2 GB of data
between an initiator system and a target system via direct I/O over an
SDR InfiniBand network (1GB/s). The block size varied between 512
bytes and 1 GB, but was always a power of two.
Expected results:
* The measurement results are consistent with the numbers I published earlier.
* During data transfers all data is transferred in blocks between 4 KB
and 32 KB in size (according to the SCST statistics).
* For small and medium block sizes (<= 32 KB) transfer times can be
modeled very well by the following formula: (transfer time) = (setup
latency) + (bytes transferred)/(bandwidth). The correlation numbers
are very close to one.
* The latency and bandwidth parameters depend on the test tool (dd
versus xdd), on the kind of test performed (reading versus writing),
on the SCSI target and on the communication protocol.
* When using RDMA (iSER or SRP), SCST has a lower latency and higher
bandwidth than STGT (results from linear regression for block sizes <=
32 KB):
Test Latency(us) Bandwidth (MB/s) Correlation
STGT+iSER, read, dd 64 560 0.999995
STGT+iSER, read, xdd 65 556 0.999994
STGT+iSER, write, dd 53 394 0.999971
STGT+iSER, write, xdd 54 445 0.999959
SCST+SRP, read, dd 39 657 0.999983
SCST+SRP, read, xdd 41 668 0.999987
SCST+SRP, write, dd 52 449 0.999962
SCST+SRP, write, xdd 52 516 0.999977
Results that I did not expect:
* A block transfer size of 1 MB is not enough to measure the maximal
throughput. The maximal throughput is only reached at much higher
block sizes (about 10 MB for SCST + SRP and about 100 MB for STGT +
iSER).
* There is one case where dd and xdd results are inconsistent: when
reading via SCST + SRP and for block sizes of about 1 MB.
* For block sizes > 64 KB the measurements differ from the model. This
is probably because all initiator-target transfers happen in blocks of
32 KB or less.
For the details and some graphs, see also
http://software.qlayer.com/display/iSCSI/Measurements .
Bart Van Assche.
On Tue, 05 Feb 2008 17:07:07 +0100
Tomasz Chmielewski <[email protected]> wrote:
> FUJITA Tomonori schrieb:
> > On Tue, 05 Feb 2008 08:14:01 +0100
> > Tomasz Chmielewski <[email protected]> wrote:
> >
> >> James Bottomley schrieb:
> >>
> >>> These are both features being independently worked on, are they not?
> >>> Even if they weren't, the combination of the size of SCST in kernel plus
> >>> the problem of having to find a migration path for the current STGT
> >>> users still looks to me to involve the greater amount of work.
> >> I don't want to be mean, but does anyone actually use STGT in
> >> production? Seriously?
> >>
> >> In the latest development version of STGT, it's only possible to stop
> >> the tgtd target daemon using KILL / 9 signal - which also means all
> >> iSCSI initiator connections are corrupted when tgtd target daemon is
> >> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
> >
> > I don't know what "iSCSI initiator connections are corrupted"
> > mean. But if you reboot a server, how can an iSCSI target
> > implementation keep iSCSI tcp connections?
>
> The problem with tgtd is that you can't start it (configured) in an
> "atomic" way.
> Usually, one will start tgtd and it's configuration in a script (I
> replaced some parameters with "..." to make it shorter and more readable):
Thanks for the details. So the way to stop the daemon is not related
with your problem.
It's easily fixable. Can you start a new thread about this on
stgt-devel mailing list? When we agree on the interface to start the
daemon, I'll implement it.
> tgtd
> tgtadm --op new ...
> tgtadm --lld iscsi --op new ...
(snip)
> So the only way to start/restart tgtd reliably is to do hacks which are
> needed with yet another iSCSI kernel implementation (IET): use iptables.
>
> iptables <block iSCSI traffic>
> tgtd
> sleep 1
> tgtadm --op new ...
> tgtadm --lld iscsi --op new ...
> iptables <unblock iSCSI traffic>
>
>
> A bit ugly, isn't it?
> Having to tinker with a firewall in order to start a daemon is by no
> means a sign of a well-tested and mature project.
>
> That's why I asked how many people use stgt in a production environment
> - James was worried about a potential migration path for current users.
I don't know how many people use stgt in a production environment but
I'm not sure that this problem prevents many people from using it in a
production environment.
You want to reboot a server running target devices while initiators
connect to it. Rebooting the target server behind the initiators
seldom works. System adminstorators in my workplace reboot storage
devices once a year and tell us to shut down the initiator machines
that use them before that.
Bart Van Assche wrote:
> On Jan 30, 2008 12:32 AM, FUJITA Tomonori <[email protected]> wrote:
>
>> iSER has parameters to limit the maximum size of RDMA (it needs to
>> repeat RDMA with a poor configuration)?
>>
>
> Please specify which parameters you are referring to. As you know I
> had already repeated my tests with ridiculously high values for the
> following iSER parameters: FirstBurstLength, MaxBurstLength and
> MaxRecvDataSegmentLength (16 MB, which is more than the 1 MB block
> size specified to dd).
>
>
Using such large values for FirstBurstLength will give you poor
performance numbers for WRITE commands (with iSER). FirstBurstLength
means how much data should you send as unsolicited data (i.e. without
RDMA). It means that your WRITE commands were sent without RDMA.
Erez
Bart Van Assche wrote:
> As you probably know there is a trend in enterprise computing towards
> networked storage. This is illustrated by the emergence during the
> past few years of standards like SRP (SCSI RDMA Protocol), iSCSI
> (Internet SCSI) and iSER (iSCSI Extensions for RDMA). Two different
> pieces of software are necessary to make networked storage possible:
> initiator software and target software. As far as I know there exist
> three different SCSI target implementations for Linux:
> - The iSCSI Enterprise Target Daemon (IETD,
> http://iscsitarget.sourceforge.net/);
> - The Linux SCSI Target Framework (STGT, http://stgt.berlios.de/);
> - The Generic SCSI Target Middle Level for Linux project (SCST,
> http://scst.sourceforge.net/).
> Since I was wondering which SCSI target software would be best suited
> for an InfiniBand network, I started evaluating the STGT and SCST SCSI
> target implementations. Apparently the performance difference between
> STGT and SCST is small on 100 Mbit/s and 1 Gbit/s Ethernet networks,
> but the SCST target software outperforms the STGT software on an
> InfiniBand network. See also the following thread for the details:
> http://sourceforge.net/mailarchive/forum.php?thread_name=e2e108260801170127w2937b2afg9bef324efa945e43%40mail.gmail.com&forum_name=scst-devel.
>
>
Sorry for the late response (but better late than never).
One may claim that STGT should have lower performance than SCST because
its data path is from userspace. However, your results show that for
non-IB transports, they both show the same numbers. Furthermore, with IB
there shouldn't be any additional difference between the 2 targets
because data transfer from userspace is as efficient as data transfer
from kernel space.
The only explanation that I see is that fine tuning for iSCSI & iSER is
required. As was already mentioned in this thread, with SDR you can get
~900 MB/sec with iSER (on STGT).
Erez
On 5-02-2008 14:38, "FUJITA Tomonori" <[email protected]> wrote:
> On Tue, 05 Feb 2008 08:14:01 +0100
> Tomasz Chmielewski <[email protected]> wrote:
>
>> James Bottomley schrieb:
>>
>>> These are both features being independently worked on, are they not?
>>> Even if they weren't, the combination of the size of SCST in kernel plus
>>> the problem of having to find a migration path for the current STGT
>>> users still looks to me to involve the greater amount of work.
>>
>> I don't want to be mean, but does anyone actually use STGT in
>> production? Seriously?
>>
>> In the latest development version of STGT, it's only possible to stop
>> the tgtd target daemon using KILL / 9 signal - which also means all
>> iSCSI initiator connections are corrupted when tgtd target daemon is
>> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
>
> I don't know what "iSCSI initiator connections are corrupted"
> mean. But if you reboot a server, how can an iSCSI target
> implementation keep iSCSI tcp connections?
>
>
>> Imagine you have to reboot all your NFS clients when you reboot your NFS
>> server. Not only that - your data is probably corrupted, or at least the
>> filesystem deserves checking...
Don't know if matters, but in my setup (iscsi on top of drbd+heartbeat)
rebooting the primary server doesn't affect my iscsi traffic, SCST correctly
manages stop/crash, by sending unit attention to clients on reconnect.
Drbd+heartbeat correctly manages those things too.
Still from an end-user POV, i was able to reboot/survive a crash only with
SCST, IETD still has reconnect problems and STGT are even worst.
Regards,
--matteo
Bart Van Assche wrote:
> On Feb 4, 2008 11:57 PM, Jeff Garzik <[email protected]> wrote:
>
>> Networked block devices are attractive because the concepts and
>> implementation are more simple than networked filesystems... but usually
>> you want to run some sort of filesystem on top. At that point you might
>> as well run NFS or [gfs|ocfs|flavor-of-the-week], and ditch your
>> networked block device (and associated complexity).
>
> Running a filesystem on top of iSCSI results in better performance
> than NFS, especially if the NFS client conforms to the NFS standard
> (=synchronous writes).
> By searching the web search for the keywords NFS, iSCSI and
> performance I found the following (6 years old) document:
> http://www.technomagesinc.com/papers/ip_paper.html. A quote from the
> conclusion:
> Our results, generated by running some of industry standard benchmarks,
> show that iSCSI significantly outperforms NFS for situations when
> performing streaming, database like accesses and small file transactions.
async performs better than sync... this is news? Furthermore, NFSv4
has not only async capability but delegation too (and RDMA if you like
such things), so the comparison is not relevant to modern times.
But a networked filesystem (note I'm using that term, not "NFS", from
here on) is simply far more useful to the average user. A networked
block device is a building block -- and a useful one. A networked
filesystem is an immediately usable solution.
For remotely accessing data, iSCSI+fs is quite simply more overhead than
a networked fs. With iSCSI you are doing
local VFS -> local blkdev -> network
whereas a networked filesystem is
local VFS -> network
iSCSI+fs also adds new manageability issues, because unless the
filesystem is single-computer (such as diskless iSCSI root fs), you
still need to go across the network _once again_ to handle filesystem
locking and coordination issues.
There is no _fundamental_ reason why remote shared storage via iSCSI OSD
is any faster than a networked filesystem.
SCSI-over-IP has its uses. Absolutely. It needed to be standardized.
But let's not pretend iSCSI is anything more than what it is. Its a
bloated cat5 cabling standard :)
Jeff
Olivier Galibert wrote:
> On Mon, Feb 04, 2008 at 05:57:47PM -0500, Jeff Garzik wrote:
>> iSCSI and NBD were passe ideas at birth. :)
>>
>> Networked block devices are attractive because the concepts and
>> implementation are more simple than networked filesystems... but usually
>> you want to run some sort of filesystem on top. At that point you might
>> as well run NFS or [gfs|ocfs|flavor-of-the-week], and ditch your
>> networked block device (and associated complexity).
>
> Call me a sysadmin, but I find easier to plug in and keep in place an
> ethernet cable than these parallel scsi cables from hell. Every
> server has at least two ethernet ports by default, with rarely any
> surprises at the kernel level. Adding ethernet cards is inexpensive,
> and you pretty much never hear of compatibility problems between
> cards.
>
> So ethernet as a connection medium is really nice compared to scsi.
> Too bad iscsi is demented and ATAoE/NBD inexistant. Maybe external
> SAS will be nice, but I don't see it getting to the level of
> universality of ethernet any time soon. And it won't get the same
> amount of user-level compatibility testing in any case.
Indeed, at the end of the day iSCSI is a bloated cabling standard. :)
It has its uses, but I don't see it as ever coming close to replacing
direct-to-network (perhaps backed with local cachefs) filesystems...
which is how all the hype comes across to me.
Cheap "Lintel" boxes everybody is familiar with _are_ the storage
appliances. Until mass-produced ATA and SCSI devices start shipping
with ethernet connectors, anyway.
Jeff
On Tue, 5 Feb 2008, Bart Van Assche wrote:
>
> Results that I did not expect:
> * A block transfer size of 1 MB is not enough to measure the maximal
> throughput. The maximal throughput is only reached at much higher
> block sizes (about 10 MB for SCST + SRP and about 100 MB for STGT +
> iSER).
Block transfer sizes over about 64kB are totally irrelevant for 99% of all
people.
Don't even bother testing anything more. Yes, bigger transfers happen, but
a lot of common loads have *smaller* transfers than 64kB.
So benchmarks that try to find "theoretical throughput" by just making big
transfers should just be banned. They give numbers, yes, but the numbers
are pointless.
Linus
This email somehow didn't manage to make it to the list (I suspect
because it had html attachments).
James
---
From:
Julian Satran
<[email protected]>
To:
Nicholas A. Bellinger
<[email protected]>
Cc:
Andrew Morton
<[email protected]>, Alan
Cox <[email protected]>, Bart
Van Assche
<[email protected]>, FUJITA
Tomonori
<[email protected]>,
James Bottomley
<[email protected]>, ...
Subject:
Re: Integration of SCST in the
mainstream Linux kernel
Date:
Mon, 4 Feb 2008 21:31:48 -0500
(20:31 CST)
Well stated. In fact the "layers" above ethernet do provide the services
that make the TCP/IP stack compelling - a whole complement of services.
ALL services required (naming, addressing, discovery, security etc.) will
have to be recreated if you take the FcOE route. That makes good business
for some but not necessary for the users. Those services BTW are not on
the data path and are not "overhead".
The TCP/IP stack pathlength is decently low. What makes most
implementations poor is that they where naively extended in the SMP world.
Recent implementations (published) from IBM and Intel show excellent
performance (4-6 times the regular stack). I do not have unfortunately
latency numbers (as the community major stress has been throughput) but I
assume that RDMA (not necessarily hardware RDMA) and/or the use of
infiniband or latency critical applications - within clusters may be the
ultimate low latency solution. Ethernet has some inherent latency issues
(the bridges) that are inherited by anything on ethernet (FcOE included).
The IP protocol stack is not inherently slow but some implementations are
somewhat sluggish.
But instead of replacing them with new and half backed contraptions we
would be all better of improving what we have and understand.
In the whole debate of around FcOE I heard a single argument that may have
some merit - building convertors iSCSI-FCP to support legacy islands of
FCP (read storage products that do not support iSCSI natively) is
expensive. It is correct technically - only that FcOE eliminates an
expense at the wrong end of the wire - it reduces the cost of the storage
box at the expense of added cost at the server (and usually there a many
servers using a storage box). FcOE vendors are also bound to provide FCP
like services for FcOE - naming, security, discovery etc. - that do not
exist on Ethernet. It is a good business for FcOE vendors - a duplicate
set of solution for users.
It should be apparent by now that if one speaks about a "converged"
network we should speak about an IP network and not about Ethernet.
If we take this route we might get perhaps also to an "infrastructure
physical variants" that support very low latency better than ethernet and
we might be able to use them with the same "stack" - a definite forward
looking solution.
IMHO it is foolish to insist on throwing away the whole stack whenever we
make a slight improvement in the physical layer of the network. We have a
substantial investment and body of knowledge in the protocol stack and
nothing proposed improves on it - obviously not as in its total level of
service nor in performance.
Julo
James Bottomley wrote:
> On Mon, 2008-02-04 at 21:38 +0300, Vladislav Bolkhovitin wrote:
>
>>James Bottomley wrote:
>>
>>>On Mon, 2008-02-04 at 20:56 +0300, Vladislav Bolkhovitin wrote:
>>>
>>>
>>>>James Bottomley wrote:
>>>>
>>>>
>>>>>On Mon, 2008-02-04 at 20:16 +0300, Vladislav Bolkhovitin wrote:
>>>>>
>>>>>
>>>>>
>>>>>>James Bottomley wrote:
>>>>>>
>>>>>>
>>>>>>
>>>>>>>>>>So, James, what is your opinion on the above? Or the overall SCSI target
>>>>>>>>>>project simplicity doesn't matter much for you and you think it's fine
>>>>>>>>>>to duplicate Linux page cache in the user space to keep the in-kernel
>>>>>>>>>>part of the project as small as possible?
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>The answers were pretty much contained here
>>>>>>>>>
>>>>>>>>>http://marc.info/?l=linux-scsi&m=120164008302435
>>>>>>>>>
>>>>>>>>>and here:
>>>>>>>>>
>>>>>>>>>http://marc.info/?l=linux-scsi&m=120171067107293
>>>>>>>>>
>>>>>>>>>Weren't they?
>>>>>>>>
>>>>>>>>No, sorry, it doesn't look so for me. They are about performance, but
>>>>>>>>I'm asking about the overall project's architecture, namely about one
>>>>>>>>part of it: simplicity. Particularly, what do you think about
>>>>>>>>duplicating Linux page cache in the user space to have zero-copy cached
>>>>>>>>I/O? Or can you suggest another architectural solution for that problem
>>>>>>>>in the STGT's approach?
>>>>>>>
>>>>>>>
>>>>>>>Isn't that an advantage of a user space solution? It simply uses the
>>>>>>>backing store of whatever device supplies the data. That means it takes
>>>>>>>advantage of the existing mechanisms for caching.
>>>>>>
>>>>>>No, please reread this thread, especially this message:
>>>>>>http://marc.info/?l=linux-kernel&m=120169189504361&w=2. This is one of
>>>>>>the advantages of the kernel space implementation. The user space
>>>>>>implementation has to have data copied between the cache and user space
>>>>>>buffer, but the kernel space one can use pages in the cache directly,
>>>>>>without extra copy.
>>>>>
>>>>>
>>>>>Well, you've said it thrice (the bellman cried) but that doesn't make it
>>>>>true.
>>>>>
>>>>>The way a user space solution should work is to schedule mmapped I/O
>>>>
>>>>>from the backing store and then send this mmapped region off for target
>>>>
>>>>>I/O. For reads, the page gather will ensure that the pages are up to
>>>>>date from the backing store to the cache before sending the I/O out.
>>>>>For writes, You actually have to do a msync on the region to get the
>>>>>data secured to the backing store.
>>>>
>>>>James, have you checked how fast is mmaped I/O if work size > size of
>>>>RAM? It's several times slower comparing to buffered I/O. It was many
>>>>times discussed in LKML and, seems, VM people consider it unavoidable.
>>>
>>>
>>>Erm, but if you're using the case of work size > size of RAM, you'll
>>>find buffered I/O won't help because you don't have the memory for
>>>buffers either.
>>
>>James, just check and you will see, buffered I/O is a lot faster.
>
> So in an out of memory situation the buffers you don't have are a lot
> faster than the pages I don't have?
There isn't OOM in both cases. Just pages reclamation/readahead work
much better in the buffered case.
>>>>So, using mmaped IO isn't an option for high performance. Plus, mmaped
>>>>IO isn't an option for high reliability requirements, since it doesn't
>>>>provide a practical way to handle I/O errors.
>>>
>>>I think you'll find it does ... the page gather returns -EFAULT if
>>>there's an I/O error in the gathered region.
>>
>>Err, to whom return? If you try to read from a mmaped page, which can't
>>be populated due to I/O error, you will get SIGBUS or SIGSEGV, I don't
>>remember exactly. It's quite tricky to get back to the faulted command
>>from the signal handler.
>>
>>Or do you mean mmap(MAP_POPULATE)/munmap() for each command? Do you
>>think that such mapping/unmapping is good for performance?
>>
>>
>>>msync does something
>>>similar if there's a write failure.
>>>
>>>
>>>>>You also have to pull tricks with
>>>>>the mmap region in the case of writes to prevent useless data being read
>>>>>in from the backing store.
>>>>
>>>>Can you be more exact and specify what kind of tricks should be done for
>>>>that?
>>>
>>>Actually, just avoid touching it seems to do the trick with a recent
>>>kernel.
>>
>>Hmm, how can one write to an mmaped page and don't touch it?
>
> I meant from user space ... the writes are done inside the kernel.
Sure, the mmap() approach agreed to be unpractical, but could you
elaborate more on this anyway, please? I'm just curious. Do you think
about implementing a new syscall, which would put pages with data in the
mmap'ed area?
> However, as Linus has pointed out, this discussion is getting a bit off
> topic.
No, that isn't off topic. We've just proved that there is no good way to
implement zero-copy cached I/O for STGT. I see the only practical way
for that, proposed by FUJITA Tomonori some time ago: duplicating Linux
page cache in the user space. But will you like it?
> There's no actual evidence that copy problems are causing any
> performatince issues issues for STGT. In fact, there's evidence that
> they're not for everything except IB networks.
The zero-copy cached I/O has not yet been implemented in SCST, I simply
so far have not had time for that. Currently SCST performs better STGT,
because of simpler processing path and less context switches per
command. Memcpy() speed on modern systems is about the same as
throughput of 20Gbps link (1600MB/s), so when the zero-copy will be
implemented, I won't be surprised by more 50-70% SCST advantage.
> James
>
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> Scst-devel mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/scst-devel
>
Linus Torvalds wrote:
> So just going by what has happened in the past, I'd assume that iSCSI
> would eventually turn into "connecting/authentication in user space" with
> "data transfers in kernel space".
This is exactly how iSCSI-SCST (iSCSI target driver for SCST) is
implemented, credits to IET and Ardis target developers.
Vlad
Linus Torvalds wrote:
>>I'd assumed the move was primarily because of the difficulty of getting
>>correct semantics on a shared filesystem
>
>
> .. not even shared. It was hard to get correct semantics full stop.
>
> Which is a traditional problem. The thing is, the kernel always has some
> internal state, and it's hard to expose all the semantics that the kernel
> knows about to user space.
>
> So no, performance is not the only reason to move to kernel space. It can
> easily be things like needing direct access to internal data queues (for a
> iSCSI target, this could be things like barriers or just tagged commands -
> yes, you can probably emulate things like that without access to the
> actual IO queues, but are you sure the semantics will be entirely right?
>
> The kernel/userland boundary is not just a performance boundary, it's an
> abstraction boundary too, and these kinds of protocols tend to break
> abstractions. NFS broke it by having "file handles" (which is not
> something that really exists in user space, and is almost impossible to
> emulate correctly), and I bet the same thing happens when emulating a SCSI
> target in user space.
Yes, there is something like that for SCSI target as well. It's a "local
initiator" or "local nexus", see
http://thread.gmane.org/gmane.linux.scsi/31288 and
http://news.gmane.org/find-root.php?message_id=%3c463F36AC.3010207%40vlnb.net%3e
for more info about that.
In fact, existence of local nexus is one more point why SCST is better,
than STGT, because for STGT it's pretty hard to support it (all locally
generated commands would have to be passed through its daemon, which
would be a total disaster for performance), while for SCST it can be
done relatively simply.
Vlad
Jeff Garzik wrote:
> Alan Cox wrote:
>
>>>better. So for example, I personally suspect that ATA-over-ethernet is way
>>>better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
>>>low-level, and against those crazy SCSI people to begin with.
>>
>>Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
>>would probably trash iSCSI for latency if nothing else.
>
>
> AoE is truly a thing of beauty. It has a two/three page RFC (say no more!).
>
> But quite so... AoE is limited to MTU size, which really hurts. Can't
> really do tagged queueing, etc.
>
>
> iSCSI is way, way too complicated.
I fully agree. From one side, all that complexity is unavoidable for
case of multiple connections per session, but for the regular case of
one connection per session it must be a lot simpler.
And now think about iSER, which brings iSCSI on the whole new complexity
level ;)
> It's an Internet protocol designed
> by storage designers, what do you expect?
>
> For years I have been hoping that someone will invent a simple protocol
> (w/ strong auth) that can transit ATA and SCSI commands and responses.
> Heck, it would be almost trivial if the kernel had a TLS/SSL implementation.
>
> Jeff
On Feb 5, 2008 6:10 PM, Erez Zilber <[email protected]> wrote:
> One may claim that STGT should have lower performance than SCST because
> its data path is from userspace. However, your results show that for
> non-IB transports, they both show the same numbers. Furthermore, with IB
> there shouldn't be any additional difference between the 2 targets
> because data transfer from userspace is as efficient as data transfer
> from kernel space.
>
> The only explanation that I see is that fine tuning for iSCSI & iSER is
> required. As was already mentioned in this thread, with SDR you can get
> ~900 MB/sec with iSER (on STGT).
My most recent measurements also show that one can get 900 MB/s with
STGT + iSER on an SDR IB network, but only for very large block sizes
(>= 100 MB). A quote from Linus Torvalds is relevant here (February 5,
2008):
Block transfer sizes over about 64kB are totally irrelevant for
99% of all people.
Please read my e-mail (posted earlier today) with a comparison for 4
KB - 64 KB block transfer sizes between SCST and STGT.
Bart Van Assche.
Erez Zilber wrote:
> Bart Van Assche wrote:
>
>>As you probably know there is a trend in enterprise computing towards
>>networked storage. This is illustrated by the emergence during the
>>past few years of standards like SRP (SCSI RDMA Protocol), iSCSI
>>(Internet SCSI) and iSER (iSCSI Extensions for RDMA). Two different
>>pieces of software are necessary to make networked storage possible:
>>initiator software and target software. As far as I know there exist
>>three different SCSI target implementations for Linux:
>>- The iSCSI Enterprise Target Daemon (IETD,
>>http://iscsitarget.sourceforge.net/);
>>- The Linux SCSI Target Framework (STGT, http://stgt.berlios.de/);
>>- The Generic SCSI Target Middle Level for Linux project (SCST,
>>http://scst.sourceforge.net/).
>>Since I was wondering which SCSI target software would be best suited
>>for an InfiniBand network, I started evaluating the STGT and SCST SCSI
>>target implementations. Apparently the performance difference between
>>STGT and SCST is small on 100 Mbit/s and 1 Gbit/s Ethernet networks,
>>but the SCST target software outperforms the STGT software on an
>>InfiniBand network. See also the following thread for the details:
>>http://sourceforge.net/mailarchive/forum.php?thread_name=e2e108260801170127w2937b2afg9bef324efa945e43%40mail.gmail.com&forum_name=scst-devel.
>>
>>
>
> Sorry for the late response (but better late than never).
>
> One may claim that STGT should have lower performance than SCST because
> its data path is from userspace. However, your results show that for
> non-IB transports, they both show the same numbers. Furthermore, with IB
> there shouldn't be any additional difference between the 2 targets
> because data transfer from userspace is as efficient as data transfer
> from kernel space.
And now consider if one target has zero-copy cached I/O. How much that
will improve its performance?
> The only explanation that I see is that fine tuning for iSCSI & iSER is
> required. As was already mentioned in this thread, with SDR you can get
> ~900 MB/sec with iSER (on STGT).
>
> Erez
Vladislav Bolkhovitin wrote:
> Jeff Garzik wrote:
>> iSCSI is way, way too complicated.
>
> I fully agree. From one side, all that complexity is unavoidable for
> case of multiple connections per session, but for the regular case of
> one connection per session it must be a lot simpler.
Actually, think about those multiple connections... we already had to
implement fast-failover (and load bal) SCSI multi-pathing at a higher
level. IMO that portion of the protocol is redundant: You need the
same capability elsewhere in the OS _anyway_, if you are to support
multi-pathing.
Jeff
On Tue, 2008-02-05 at 21:59 +0300, Vladislav Bolkhovitin wrote:
> >>Hmm, how can one write to an mmaped page and don't touch it?
> >
> > I meant from user space ... the writes are done inside the kernel.
>
> Sure, the mmap() approach agreed to be unpractical, but could you
> elaborate more on this anyway, please? I'm just curious. Do you think
> about implementing a new syscall, which would put pages with data in the
> mmap'ed area?
No, it has to do with the way invalidation occurs. When you mmap a
region from a device or file, the kernel places page translations for
that region into your vm_area. The regions themselves aren't backed
until faulted. For write (i.e. incoming command to target) you specify
the write flag and send the area off to receive the data. The gather,
expecting the pages to be overwritten, backs them with pages marked
dirty but doesn't fault in the contents (unless it already exists in the
page cache). The kernel writes the data to the pages and the dirty
pages go back to the user. msync() flushes them to the device.
The disadvantage of all this is that the handle for the I/O if you will
is a virtual address in a user process that doesn't actually care to see
the data. non-x86 architectures will do flushes/invalidates on this
address space as the I/O occurs.
> > However, as Linus has pointed out, this discussion is getting a bit off
> > topic.
>
> No, that isn't off topic. We've just proved that there is no good way to
> implement zero-copy cached I/O for STGT. I see the only practical way
> for that, proposed by FUJITA Tomonori some time ago: duplicating Linux
> page cache in the user space. But will you like it?
Well, there's no real evidence that zero copy or lack of it is a problem
yet.
James
Jeff Garzik wrote:
>>> iSCSI is way, way too complicated.
>>
>> I fully agree. From one side, all that complexity is unavoidable for
>> case of multiple connections per session, but for the regular case of
>> one connection per session it must be a lot simpler.
>
> Actually, think about those multiple connections... we already had to
> implement fast-failover (and load bal) SCSI multi-pathing at a higher
> level. IMO that portion of the protocol is redundant: You need the
> same capability elsewhere in the OS _anyway_, if you are to support
> multi-pathing.
I'm thinking about MC/S as about a way to improve performance using
several physical links. There's no other way, except MC/S, to keep
commands processing order in that case. So, it's really valuable
property of iSCSI, although with a limited application.
Vlad
On Tue, 2008-02-05 at 22:21 +0300, Vladislav Bolkhovitin wrote:
> Jeff Garzik wrote:
> >>> iSCSI is way, way too complicated.
> >>
> >> I fully agree. From one side, all that complexity is unavoidable for
> >> case of multiple connections per session, but for the regular case of
> >> one connection per session it must be a lot simpler.
> >
> > Actually, think about those multiple connections... we already had to
> > implement fast-failover (and load bal) SCSI multi-pathing at a higher
> > level. IMO that portion of the protocol is redundant: You need the
> > same capability elsewhere in the OS _anyway_, if you are to support
> > multi-pathing.
>
> I'm thinking about MC/S as about a way to improve performance using
> several physical links. There's no other way, except MC/S, to keep
> commands processing order in that case. So, it's really valuable
> property of iSCSI, although with a limited application.
>
> Vlad
>
Greetings,
I have always observed the case with LIO SE/iSCSI target mode (as well
as with other software initiators we can leave out of the discussion for
now, and congrats to the open/iscsi on folks recent release. :-) that
execution core hardware thread and inter-nexus per 1 Gb/sec ethernet
port performance scales up to 4x and 2x core x86_64 very well with
MC/S). I have been seeing 450 MB/sec using 2x socket 4x core x86_64 for
a number of years with MC/S. Using MC/S on 10 Gb/sec (on PCI-X v2.0
266mhz as well, which was the first transport that LIO Target ran on
that was able to reach handle duplex ~1200 MB/sec with 3 initiators and
MC/S. In the point to point 10 GB/sec tests on IBM p404 machines, the
initiators where able to reach ~910 MB/sec with MC/S. Open/iSCSI was
able to go a bit faster (~950 MB/sec) because it uses struct sk_buff
directly.
A good rule to keep in mind here while considering performance is that
context switching overhead and pipeline <-> bus stalling (along with
other legacy OS specific storage stack limitations with BLOCK and VFS
with O_DIRECT, et al and I will leave out of the discussion for iSCSI
and SE engine target mode) is that a initiator will scale roughly 1/2 as
well as a target, given comparable hardware and virsh output. The
software target case target case also depends, in great regard in many
cases, if we are talking about something something as simple as doing
contiguous DMA memory allocations in from a SINGLE kernel thread, and
handling direction execution to a storage hardware DMA ring that may
have not been allocated in the current kernel thread. In MC/S mode this
breaks down to:
1) Sorting logic that handles pre execution statemachine for transport
from local RDMA memory and OS specific data buffers. TCP application
data buffer, struct sk_buff, or RDMA struct page or SG. This should be
generic between iSCSI and iSER.
2) Allocation of said memory buffers to OS subsystem dependent code that
can be queued up to these drivers. It breaks down to what you can get
drivers and OS subsystem folks to agree to implement, and can be made
generic in a Transport / BLOCK / VFS layered storage stack. In the
"allocate thread DMA ring and use OS supported software and vendor
available hardware" I don't think the kernel space requirement will
every completely be able to go away.
Without diving into RFC-3720 specifics, the statemachine for MC/S side
for memory allocation, login and logout generic to iSCSi and ISER, and
ERL=2 recovery. My plan is to post the locations in the LIO code where
this has been implemented, and where we where can make this easier, etc.
In the early in the development of what eventually became LIO Target
code, ERL was broken into separete files and separete function
prefixes.
iscsi_target_erl0, iscsi_target_erl1 and iscsi_target_erl2.
The statemachine for ERL=0 and ERL=2 is pretty simple in RFC-3720 (have
a look for those interested in the discussion)
7.1.1. State Descriptions for Initiators and Targets
The LIO target code is also pretty simple for this:
[root@ps3-cell target]# wc -l iscsi_target_erl*
1115 iscsi_target_erl0.c
45 iscsi_target_erl0.h
526 iscsi_target_erl0.o
1426 iscsi_target_erl1.c
51 iscsi_target_erl1.h
1253 iscsi_target_erl1.o
605 iscsi_target_erl2.c
45 iscsi_target_erl2.h
447 iscsi_target_erl2.o
5513 total
erl1.c is a bit larger than the others because it contains the MC/S
statemachine functions. iscsi_target_erl1.c:iscsi_execute_cmd() and
iscsi_target_util.c:iscsi_check_received_cmdsn() do most of the work for
LIO MC/S state machine. I would probably benefit from being in broken
up into say iscsi_target_mcs.c. Note that all of this code is MC/S
safe, with the exception of the specific SCSI TMR functions. For the
SCSI TMR pieces, I have always hoped to use SCST code for doing this...
Most of the login/logout code is done in iscsi_target.c, which is could
probably also benefit fot getting broken out...
--nab
On Tue, 2008-02-05 at 14:12 -0500, Jeff Garzik wrote:
> Vladislav Bolkhovitin wrote:
> > Jeff Garzik wrote:
> >> iSCSI is way, way too complicated.
> >
> > I fully agree. From one side, all that complexity is unavoidable for
> > case of multiple connections per session, but for the regular case of
> > one connection per session it must be a lot simpler.
>
>
> Actually, think about those multiple connections... we already had to
> implement fast-failover (and load bal) SCSI multi-pathing at a higher
> level. IMO that portion of the protocol is redundant: You need the
> same capability elsewhere in the OS _anyway_, if you are to support
> multi-pathing.
>
> Jeff
>
>
Hey Jeff,
I put a whitepaper on the LIO cluster recently about this topic.. It is
from a few years ago but the datapoints are very relevant.
http://linux-iscsi.org/builds/user/nab/Inter.vs.OuterNexus.Multiplexing.pdf
The key advantage to MC/S and ERL=2 has always been that they are
completely OS independent. They are designed to work together and
actually benefit from one another.
They are also are protocol independent between Traditional iSCSI and
iSER.
--nab
PS: A great thanks for my former colleague Edward Cheng for putting this
together.
>
>
On Tue, 2008-02-05 at 22:01 +0300, Vladislav Bolkhovitin wrote:
> Jeff Garzik wrote:
> > Alan Cox wrote:
> >
> >>>better. So for example, I personally suspect that ATA-over-ethernet is way
> >>>better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> >>>low-level, and against those crazy SCSI people to begin with.
> >>
> >>Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> >>would probably trash iSCSI for latency if nothing else.
> >
> >
> > AoE is truly a thing of beauty. It has a two/three page RFC (say no more!).
> >
> > But quite so... AoE is limited to MTU size, which really hurts. Can't
> > really do tagged queueing, etc.
> >
> >
> > iSCSI is way, way too complicated.
>
> I fully agree. From one side, all that complexity is unavoidable for
> case of multiple connections per session, but for the regular case of
> one connection per session it must be a lot simpler.
>
> And now think about iSER, which brings iSCSI on the whole new complexity
> level ;)
Actually, the iSER protocol wire protocol itself is quite simple,
because it builds on iSCSI and IPS fundamentals, and because traditional
iSCSI's recovery logic for CRC failures (and hence alot of
acknowledgement sequence PDUs that go missing, etc) and the RDMA Capable
Protocol (RCaP).
The logic that iSER collectively disables is known as within-connection
and within-command recovery (negotiated as ErrorRecoveryLevel=1 on the
wire), RFC-5046 requires that the iSCSI layer that iSER is being enabled
to disable CRC32C checksums and any associated timeouts for ERL=1.
Also, have a look at Appendix A. in the iSER spec.
A.1. iWARP Message Format for iSER Hello Message ...............73
A.2. iWARP Message Format for iSER HelloReply Message ..........74
A.3. iWARP Message Format for SCSI Read Command PDU ............75
A.4. iWARP Message Format for SCSI Read Data ...................76
A.5. iWARP Message Format for SCSI Write Command PDU ...........77
A.6. iWARP Message Format for RDMA Read Request ................78
A.7. iWARP Message Format for Solicited SCSI Write Data ........79
A.8. iWARP Message Format for SCSI Response PDU ................80
This is about as 1/2 as many traditional iSCSI PDUs, that iSER
encapulates.
--nab
On Tue, 2008-02-05 at 16:48 -0800, Nicholas A. Bellinger wrote:
> On Tue, 2008-02-05 at 22:01 +0300, Vladislav Bolkhovitin wrote:
> > Jeff Garzik wrote:
> > > Alan Cox wrote:
> > >
> > >>>better. So for example, I personally suspect that ATA-over-ethernet is way
> > >>>better than some crazy SCSI-over-TCP crap, but I'm biased for simple and
> > >>>low-level, and against those crazy SCSI people to begin with.
> > >>
> > >>Current ATAoE isn't. It can't support NCQ. A variant that did NCQ and IP
> > >>would probably trash iSCSI for latency if nothing else.
> > >
> > >
> > > AoE is truly a thing of beauty. It has a two/three page RFC (say no more!).
> > >
> > > But quite so... AoE is limited to MTU size, which really hurts. Can't
> > > really do tagged queueing, etc.
> > >
> > >
> > > iSCSI is way, way too complicated.
> >
> > I fully agree. From one side, all that complexity is unavoidable for
> > case of multiple connections per session, but for the regular case of
> > one connection per session it must be a lot simpler.
> >
> > And now think about iSER, which brings iSCSI on the whole new complexity
> > level ;)
>
> Actually, the iSER protocol wire protocol itself is quite simple,
> because it builds on iSCSI and IPS fundamentals, and because traditional
> iSCSI's recovery logic for CRC failures (and hence alot of
> acknowledgement sequence PDUs that go missing, etc) and the RDMA
> Capable
> Protocol (RCaP).
this should be:
.. and instead the RDMA Capacle Protocol (RCaP) provides the 32-bit or
greater data integrity.
--nab
On Tue, 05 Feb 2008 18:09:15 +0100
Matteo Tescione <[email protected]> wrote:
> On 5-02-2008 14:38, "FUJITA Tomonori" <[email protected]> wrote:
>
> > On Tue, 05 Feb 2008 08:14:01 +0100
> > Tomasz Chmielewski <[email protected]> wrote:
> >
> >> James Bottomley schrieb:
> >>
> >>> These are both features being independently worked on, are they not?
> >>> Even if they weren't, the combination of the size of SCST in kernel plus
> >>> the problem of having to find a migration path for the current STGT
> >>> users still looks to me to involve the greater amount of work.
> >>
> >> I don't want to be mean, but does anyone actually use STGT in
> >> production? Seriously?
> >>
> >> In the latest development version of STGT, it's only possible to stop
> >> the tgtd target daemon using KILL / 9 signal - which also means all
> >> iSCSI initiator connections are corrupted when tgtd target daemon is
> >> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
> >
> > I don't know what "iSCSI initiator connections are corrupted"
> > mean. But if you reboot a server, how can an iSCSI target
> > implementation keep iSCSI tcp connections?
> >
> >
> >> Imagine you have to reboot all your NFS clients when you reboot your NFS
> >> server. Not only that - your data is probably corrupted, or at least the
> >> filesystem deserves checking...
>
> Don't know if matters, but in my setup (iscsi on top of drbd+heartbeat)
> rebooting the primary server doesn't affect my iscsi traffic, SCST correctly
> manages stop/crash, by sending unit attention to clients on reconnect.
> Drbd+heartbeat correctly manages those things too.
> Still from an end-user POV, i was able to reboot/survive a crash only with
> SCST, IETD still has reconnect problems and STGT are even worst.
Please tell us on stgt-devel mailing list if you see problems. We will
try to fix them.
Thanks,
On Tue, 2008-02-05 at 16:11 -0800, Nicholas A. Bellinger wrote:
> On Tue, 2008-02-05 at 22:21 +0300, Vladislav Bolkhovitin wrote:
> > Jeff Garzik wrote:
> > >>> iSCSI is way, way too complicated.
> > >>
> > >> I fully agree. From one side, all that complexity is unavoidable for
> > >> case of multiple connections per session, but for the regular case of
> > >> one connection per session it must be a lot simpler.
> > >
> > > Actually, think about those multiple connections... we already had to
> > > implement fast-failover (and load bal) SCSI multi-pathing at a higher
> > > level. IMO that portion of the protocol is redundant: You need the
> > > same capability elsewhere in the OS _anyway_, if you are to support
> > > multi-pathing.
> >
> > I'm thinking about MC/S as about a way to improve performance using
> > several physical links. There's no other way, except MC/S, to keep
> > commands processing order in that case. So, it's really valuable
> > property of iSCSI, although with a limited application.
> >
> > Vlad
> >
>
> Greetings,
>
> I have always observed the case with LIO SE/iSCSI target mode (as well
> as with other software initiators we can leave out of the discussion for
> now, and congrats to the open/iscsi on folks recent release. :-) that
> execution core hardware thread and inter-nexus per 1 Gb/sec ethernet
> port performance scales up to 4x and 2x core x86_64 very well with
> MC/S). I have been seeing 450 MB/sec using 2x socket 4x core x86_64 for
> a number of years with MC/S. Using MC/S on 10 Gb/sec (on PCI-X v2.0
> 266mhz as well, which was the first transport that LIO Target ran on
> that was able to reach handle duplex ~1200 MB/sec with 3 initiators and
> MC/S. In the point to point 10 GB/sec tests on IBM p404 machines, the
> initiators where able to reach ~910 MB/sec with MC/S. Open/iSCSI was
> able to go a bit faster (~950 MB/sec) because it uses struct sk_buff
> directly.
>
Sorry, these where IBM p505 express (not p404, duh) which had a 2x
socket 2x core POWER5 setup. These along with an IBM X-series machine)
where the only ones available for PCI-X v2.0, and this probably is still
the case. :-)
Also, these numbers where with a ~9000 MTU (I don't recall what the
hardware limit on the 10 Gb/sec switch lwas) doing direct struct iovec
to preallocated struct page mapping for payload on the target side.
This is known as RAMDISK_DR plugin in the LIO-SE. On the initiator, LTP
disktest and O_DIRECT where used for direct to SCSI block device access.
I can big up this paper if anyone is interested.
--nab
On Wed, 2008-02-06 at 10:29 +0900, FUJITA Tomonori wrote:
> On Tue, 05 Feb 2008 18:09:15 +0100
> Matteo Tescione <[email protected]> wrote:
>
> > On 5-02-2008 14:38, "FUJITA Tomonori" <[email protected]> wrote:
> >
> > > On Tue, 05 Feb 2008 08:14:01 +0100
> > > Tomasz Chmielewski <[email protected]> wrote:
> > >
> > >> James Bottomley schrieb:
> > >>
> > >>> These are both features being independently worked on, are they not?
> > >>> Even if they weren't, the combination of the size of SCST in kernel plus
> > >>> the problem of having to find a migration path for the current STGT
> > >>> users still looks to me to involve the greater amount of work.
> > >>
> > >> I don't want to be mean, but does anyone actually use STGT in
> > >> production? Seriously?
> > >>
> > >> In the latest development version of STGT, it's only possible to stop
> > >> the tgtd target daemon using KILL / 9 signal - which also means all
> > >> iSCSI initiator connections are corrupted when tgtd target daemon is
> > >> started again (kernel upgrade, target daemon upgrade, server reboot etc.).
> > >
> > > I don't know what "iSCSI initiator connections are corrupted"
> > > mean. But if you reboot a server, how can an iSCSI target
> > > implementation keep iSCSI tcp connections?
> > >
> > >
> > >> Imagine you have to reboot all your NFS clients when you reboot your NFS
> > >> server. Not only that - your data is probably corrupted, or at least the
> > >> filesystem deserves checking...
> >
The TCP connection will drop, remember that the TCP connection state for
one side has completely vanished. Depending on iSCSI/iSER
ErrorRecoveryLevel that is set, this will mean:
1) Session Recovery, ERL=0 - Restarting the entire nexus and all
connections across all of the possible subnets or comm-links. All
outstanding un-StatSN acknowledged commands will be returned back to the
SCSI subsystem with RETRY status. Once a single connection has been
reestablished to start the nexus, the CDBs will be resent.
2) Connection Recovery, ERL=2 - CDBs from the failed connection(s) will
be retried (nothing changes in the PDU) to fill the iSCSI CmdSN ordering
gap, or be explictly retried with TMR TASK_REASSIGN for ones already
acknowledged by the ExpCmdSN that are returned to the initiator in
response packets or by way of unsolicited NopINs.
> > Don't know if matters, but in my setup (iscsi on top of drbd+heartbeat)
> > rebooting the primary server doesn't affect my iscsi traffic, SCST correctly
> > manages stop/crash, by sending unit attention to clients on reconnect.
> > Drbd+heartbeat correctly manages those things too.
> > Still from an end-user POV, i was able to reboot/survive a crash only with
> > SCST, IETD still has reconnect problems and STGT are even worst.
>
> Please tell us on stgt-devel mailing list if you see problems. We will
> try to fix them.
>
FYI, the LIO code also supports rmmoding iscsi_target_mod while at full
10 Gb/sec speed. I think it should be a requirement to be able to
control per initiator, per portal group, per LUN, per device, per HBA in
the design without restarting any other objects.
--nab
> Thanks,
> --
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at http://www.tux.org/lkml/
>
On Feb 5, 2008 6:50 PM, Jeff Garzik <[email protected]> wrote:
> For remotely accessing data, iSCSI+fs is quite simply more overhead than
> a networked fs. With iSCSI you are doing
>
> local VFS -> local blkdev -> network
>
> whereas a networked filesystem is
>
> local VFS -> network
There are use cases than can be solved better via iSCSI and a
filesystem than via a network filesystem. One such use case is when
deploying a virtual machine whose data is stored on a network server:
in that case there is only one user of the data (so there are no
locking issues) and filesystem and block device each run in another
operating system: the filesystem runs inside the virtual machine and
iSCSI either runs in the hypervisor or in the native OS.
Bart Van Assche.
On Feb 5, 2008 6:01 PM, Erez Zilber <[email protected]> wrote:
>
> Using such large values for FirstBurstLength will give you poor
> performance numbers for WRITE commands (with iSER). FirstBurstLength
> means how much data should you send as unsolicited data (i.e. without
> RDMA). It means that your WRITE commands were sent without RDMA.
Sorry, but I'm afraid you got this wrong. When the iSER transport is
used instead of TCP, all data is sent via RDMA, including unsolicited
data. If you have look at the iSER implementation in the Linux kernel
(source files under drivers/infiniband/ulp/iser), you will see that
all data is transferred via RDMA and not via TCP/IP.
Bart Van Assche.
Bart Van Assche wrote:
> On Feb 5, 2008 6:50 PM, Jeff Garzik <[email protected]> wrote:
>> For remotely accessing data, iSCSI+fs is quite simply more overhead than
>> a networked fs. With iSCSI you are doing
>>
>> local VFS -> local blkdev -> network
>>
>> whereas a networked filesystem is
>>
>> local VFS -> network
>
> There are use cases than can be solved better via iSCSI and a
> filesystem than via a network filesystem. One such use case is when
> deploying a virtual machine whose data is stored on a network server:
> in that case there is only one user of the data (so there are no
> locking issues) and filesystem and block device each run in another
> operating system: the filesystem runs inside the virtual machine and
> iSCSI either runs in the hypervisor or in the native OS.
Hence the diskless root fs configuration I referred to in multiple
emails... whoopee, you reinvented NFS root with quotas :)
Jeff
On Feb. 06, 2008, 14:16 +0200, "Bart Van Assche" <[email protected]> wrote:
> On Feb 5, 2008 6:01 PM, Erez Zilber <[email protected]> wrote:
>> Using such large values for FirstBurstLength will give you poor
>> performance numbers for WRITE commands (with iSER). FirstBurstLength
>> means how much data should you send as unsolicited data (i.e. without
>> RDMA). It means that your WRITE commands were sent without RDMA.
>
> Sorry, but I'm afraid you got this wrong. When the iSER transport is
> used instead of TCP, all data is sent via RDMA, including unsolicited
> data. If you have look at the iSER implementation in the Linux kernel
> (source files under drivers/infiniband/ulp/iser), you will see that
> all data is transferred via RDMA and not via TCP/IP.
Regardless of what the current implementation is, the behavior you (Bart)
describe seems to disagree with http://www.ietf.org/rfc/rfc5046.txt.
Benny
>
> Bart Van Assche.
> -
> Sorry, but I'm afraid you got this wrong. When the iSER transport is
> used instead of TCP, all data is sent via RDMA, including unsolicited
> data. If you have look at the iSER implementation in the Linux kernel
> (source files under drivers/infiniband/ulp/iser), you will see that
> all data is transferred via RDMA and not via TCP/IP.
I think the confusion here is caused by a slight misuse of the term
"RDMA". It is true that all data is always transported over an
InfiniBand connection when iSER is used, but not all such transfers
are one-sided RDMA operations; some data can be transferred using
send/receive operations.
- R.
James Bottomley wrote:
> On Tue, 2008-02-05 at 21:59 +0300, Vladislav Bolkhovitin wrote:
>
>>>>Hmm, how can one write to an mmaped page and don't touch it?
>>>
>>>I meant from user space ... the writes are done inside the kernel.
>>
>>Sure, the mmap() approach agreed to be unpractical, but could you
>>elaborate more on this anyway, please? I'm just curious. Do you think
>>about implementing a new syscall, which would put pages with data in the
>>mmap'ed area?
>
> No, it has to do with the way invalidation occurs. When you mmap a
> region from a device or file, the kernel places page translations for
> that region into your vm_area. The regions themselves aren't backed
> until faulted. For write (i.e. incoming command to target) you specify
> the write flag and send the area off to receive the data. The gather,
> expecting the pages to be overwritten, backs them with pages marked
> dirty but doesn't fault in the contents (unless it already exists in the
> page cache). The kernel writes the data to the pages and the dirty
> pages go back to the user. msync() flushes them to the device.
>
> The disadvantage of all this is that the handle for the I/O if you will
> is a virtual address in a user process that doesn't actually care to see
> the data. non-x86 architectures will do flushes/invalidates on this
> address space as the I/O occurs.
I more or less see, thanks. But (1) pages still needs to be mmaped to
the user space process before the data transmission, i.e. they must be
zeroed before being mmaped, which isn't much faster, than data copy, and
(2) I suspect, it would be hard to make it race free, e.g. if another
process would want to write to the same area simultaneously
>>>However, as Linus has pointed out, this discussion is getting a bit off
>>>topic.
>>
>>No, that isn't off topic. We've just proved that there is no good way to
>>implement zero-copy cached I/O for STGT. I see the only practical way
>>for that, proposed by FUJITA Tomonori some time ago: duplicating Linux
>>page cache in the user space. But will you like it?
>
> Well, there's no real evidence that zero copy or lack of it is a problem
> yet.
The performance improvement from zero copy can be easily estimated,
knowing the link throughput and data copy throughput, which are about
the same for 20Gbps links (I did that few e-mail ago).
Vlad
Since the focus of this thread shifted somewhat in the last few
messages, I'll try to summarize what has been discussed so far:
- There was a number of participants who joined this discussion
spontaneously. This suggests that there is considerable interest in
networked storage and iSCSI.
- It has been motivated why iSCSI makes sense as a storage protocol
(compared to ATA over Ethernet and Fibre Channel over Ethernet).
- The direct I/O performance results for block transfer sizes below 64
KB are a meaningful benchmark for storage target implementations.
- It has been discussed whether an iSCSI target should be implemented
in user space or in kernel space. It is clear now that an
implementation in the kernel can be made faster than a user space
implementation (http://kerneltrap.org/mailarchive/linux-kernel/2008/2/4/714804).
Regarding existing implementations, measurements have a.o. shown that
SCST is faster than STGT (30% with the following setup: iSCSI via
IPoIB and direct I/O block transfers with a size of 512 bytes).
- It has been discussed which iSCSI target implementation should be in
the mainstream Linux kernel. There is no agreement on this subject
yet. The short-term options are as follows:
1) Do not integrate any new iSCSI target implementation in the
mainstream Linux kernel.
2) Add one of the existing in-kernel iSCSI target implementations to
the kernel, e.g. SCST or PyX/LIO.
3) Create a new in-kernel iSCSI target implementation that combines
the advantages of the existing iSCSI kernel target implementations
(iETD, STGT, SCST and PyX/LIO).
As an iSCSI user, I prefer option (3). The big question is whether the
various storage target authors agree with this ?
Bart Van Assche.
Bart Van Assche wrote:
> Since the focus of this thread shifted somewhat in the last few
> messages, I'll try to summarize what has been discussed so far:
> - There was a number of participants who joined this discussion
> spontaneously. This suggests that there is considerable interest in
> networked storage and iSCSI.
> - It has been motivated why iSCSI makes sense as a storage protocol
> (compared to ATA over Ethernet and Fibre Channel over Ethernet).
> - The direct I/O performance results for block transfer sizes below 64
> KB are a meaningful benchmark for storage target implementations.
> - It has been discussed whether an iSCSI target should be implemented
> in user space or in kernel space. It is clear now that an
> implementation in the kernel can be made faster than a user space
> implementation (http://kerneltrap.org/mailarchive/linux-kernel/2008/2/4/714804).
> Regarding existing implementations, measurements have a.o. shown that
> SCST is faster than STGT (30% with the following setup: iSCSI via
> IPoIB and direct I/O block transfers with a size of 512 bytes).
> - It has been discussed which iSCSI target implementation should be in
> the mainstream Linux kernel. There is no agreement on this subject
> yet. The short-term options are as follows:
> 1) Do not integrate any new iSCSI target implementation in the
> mainstream Linux kernel.
> 2) Add one of the existing in-kernel iSCSI target implementations to
> the kernel, e.g. SCST or PyX/LIO.
> 3) Create a new in-kernel iSCSI target implementation that combines
> the advantages of the existing iSCSI kernel target implementations
> (iETD, STGT, SCST and PyX/LIO).
>
> As an iSCSI user, I prefer option (3). The big question is whether the
> various storage target authors agree with this ?
I tend to agree with some important notes:
1. IET should be excluded from this list, iSCSI-SCST is IET updated for
SCST framework with a lot of bugfixes and improvements.
2. I think, everybody will agree that Linux iSCSI target should work
over some standard SCSI target framework. Hence the choice gets
narrower: SCST vs STGT. I don't think there's a way for a dedicated
iSCSI target (i.e. PyX/LIO) in the mainline, because of a lot of code
duplication. Nicholas could decide to move to either existing framework
(although, frankly, I don't think there's a possibility for in-kernel
iSCSI target and user space SCSI target framework) and if he decide to
go with SCST, I'll be glad to offer my help and support and wouldn't
care if LIO-SCST eventually replaced iSCSI-SCST. The better one should win.
Vlad
On Thu, 2008-02-07 at 14:13 +0100, Bart Van Assche wrote:
> Since the focus of this thread shifted somewhat in the last few
> messages, I'll try to summarize what has been discussed so far:
> - There was a number of participants who joined this discussion
> spontaneously. This suggests that there is considerable interest in
> networked storage and iSCSI.
> - It has been motivated why iSCSI makes sense as a storage protocol
> (compared to ATA over Ethernet and Fibre Channel over Ethernet).
> - The direct I/O performance results for block transfer sizes below 64
> KB are a meaningful benchmark for storage target implementations.
> - It has been discussed whether an iSCSI target should be implemented
> in user space or in kernel space. It is clear now that an
> implementation in the kernel can be made faster than a user space
> implementation (http://kerneltrap.org/mailarchive/linux-kernel/2008/2/4/714804).
> Regarding existing implementations, measurements have a.o. shown that
> SCST is faster than STGT (30% with the following setup: iSCSI via
> IPoIB and direct I/O block transfers with a size of 512 bytes).
> - It has been discussed which iSCSI target implementation should be in
> the mainstream Linux kernel. There is no agreement on this subject
> yet. The short-term options are as follows:
> 1) Do not integrate any new iSCSI target implementation in the
> mainstream Linux kernel.
> 2) Add one of the existing in-kernel iSCSI target implementations to
> the kernel, e.g. SCST or PyX/LIO.
> 3) Create a new in-kernel iSCSI target implementation that combines
> the advantages of the existing iSCSI kernel target implementations
> (iETD, STGT, SCST and PyX/LIO).
>
> As an iSCSI user, I prefer option (3). The big question is whether the
> various storage target authors agree with this ?
>
I think the other data point here would be that final target design
needs to be as generic as possible. Generic in the sense that the
engine eventually needs to be able to accept NDB and other ethernet
based target mode storage configurations to an abstracted device object
(struct scsi_device, struct block_device, or struct file) just as it
would for an IP Storage based request.
We know that NDB and *oE will have their own naming and discovery, and
the first set of IO tasks to be completed would be those using
(iscsi_cmd_t->cmd_flags & ICF_SCSI_DATA_SG_IO_CDB) in
iscsi_target_transport.c in the current code. These are single READ_*
and WRITE_* codepaths that perform DMA memory pre-proceessing in v2.9
LIO-SE.
Also, being able to tell the engine to accelerate to DMA ring operation
(say to underlying struct scsi_device or struct block_device) instead of
fileio in some cases you will see better performance when using hardware
(ie: not a underlying kernel thread queueing IO into block). But I have
found FILEIO with sendpage with MD to be faster in single threaded tests
than struct block_device. I am currently using IBLOCK for LVM for core
LIO operation (which actually sits on software MD raid6). I do this
because using submit_bio() with se_mem_t mapped arrays of struct
scatterlist -> struct bio_vec can handle power failures properly, and
not send back StatSN Acks to the Initiator who thinks that everything
has already made it to disk. This is the case with doing IO to struct
file in the kernel today without a kernel level O_DIRECT.
Also for proper kernel-level target mode support, using struct file with
O_DIRECT for storage blocks and emulating control path CDBS is one of
the work items. This can be made generic or obtained from the
underlying storage object (anything that can be exported from LIO
Subsystem TPI) For real hardware (struct scsi_device in just about all
the cases these days). Last time I looked this was due to
fs/direct-io.c:dio_refill_pages() using get_user_pages()...
For really transport specific CDB and control code, which in good amount
of cases, we are going eventually be expected to emulate in software.
I really like how STGT breaks this up into per device type code
segments; spc.c sbc.c mmc.c ssc.c smc.c etc. Having all of these split
out properly is one strong point of STGT IMHO, and really makes learning
things much easier. Also, being able to queue these IOs into a
userspace and receive a asynchronous response back up the storage stack.
I think this is actually a pretty interesting potential for passing
storage protocol packets into userspace apps and leave the protocol
state machines and recovery paths in the kernel with a generic target
engine.
Also, I know that the SCST folks have put alot of time into getting the
very SCSI hardware specific target mode control modes to work. I
personally own a bunch of this adapters, and would really like to see
better support for target mode on non iSCSI type adapters with a single
target mode storage engine that abstracts storage subsystems and wire
protocol fabrics.
--nab
Is there an open iSCSI Target implementation which does NOT
issue commands to sub-target devices via the SCSI mid-layer, but
bypasses it completely?
Luben
On Thu, 7 Feb 2008, Vladislav Bolkhovitin wrote:
> Bart Van Assche wrote:
>> - It has been discussed which iSCSI target implementation should be in
>> the mainstream Linux kernel. There is no agreement on this subject
>> yet. The short-term options are as follows:
>> 1) Do not integrate any new iSCSI target implementation in the
>> mainstream Linux kernel.
>> 2) Add one of the existing in-kernel iSCSI target implementations to
>> the kernel, e.g. SCST or PyX/LIO.
>> 3) Create a new in-kernel iSCSI target implementation that combines
>> the advantages of the existing iSCSI kernel target implementations
>> (iETD, STGT, SCST and PyX/LIO).
>>
>> As an iSCSI user, I prefer option (3). The big question is whether the
>> various storage target authors agree with this ?
>
> I tend to agree with some important notes:
>
> 1. IET should be excluded from this list, iSCSI-SCST is IET updated for SCST
> framework with a lot of bugfixes and improvements.
>
> 2. I think, everybody will agree that Linux iSCSI target should work over
> some standard SCSI target framework. Hence the choice gets narrower: SCST vs
> STGT. I don't think there's a way for a dedicated iSCSI target (i.e. PyX/LIO)
> in the mainline, because of a lot of code duplication. Nicholas could decide
> to move to either existing framework (although, frankly, I don't think
> there's a possibility for in-kernel iSCSI target and user space SCSI target
> framework) and if he decide to go with SCST, I'll be glad to offer my help
> and support and wouldn't care if LIO-SCST eventually replaced iSCSI-SCST. The
> better one should win.
why should linux as an iSCSI target be limited to passthrough to a SCSI
device.
the most common use of this sort of thing that I would see is to load up a
bunch of 1TB SATA drives in a commodity PC, run software RAID, and then
export the resulting volume to other servers via iSCSI. not a 'real' SCSI
device in sight.
As far as how good a standard iSCSI is, at this point I don't think it
really matters. There are too many devices and manufacturers out there
that implement iSCSI as their storage protocol (from both sides, offering
storage to other systems, and using external storage). Sometimes the best
technology doesn't win, but Linux should be interoperable with as much as
possible and be ready to support the winners and the loosers in technology
options, for as long as anyone chooses to use the old equipment (after
all, we support things like Arcnet networking, which lost to Ethernet many
years ago)
David Lang
Luben Tuikov wrote:
> Is there an open iSCSI Target implementation which does NOT
> issue commands to sub-target devices via the SCSI mid-layer, but
> bypasses it completely?
What do you mean? To call directly low level backstorage SCSI drivers
queuecommand() routine? What are advantages of it?
> Luben
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
[email protected] wrote:
> On Thu, 7 Feb 2008, Vladislav Bolkhovitin wrote:
>
>> Bart Van Assche wrote:
>>
>>> - It has been discussed which iSCSI target implementation should be in
>>> the mainstream Linux kernel. There is no agreement on this subject
>>> yet. The short-term options are as follows:
>>> 1) Do not integrate any new iSCSI target implementation in the
>>> mainstream Linux kernel.
>>> 2) Add one of the existing in-kernel iSCSI target implementations to
>>> the kernel, e.g. SCST or PyX/LIO.
>>> 3) Create a new in-kernel iSCSI target implementation that combines
>>> the advantages of the existing iSCSI kernel target implementations
>>> (iETD, STGT, SCST and PyX/LIO).
>>>
>>> As an iSCSI user, I prefer option (3). The big question is whether the
>>> various storage target authors agree with this ?
>>
>>
>> I tend to agree with some important notes:
>>
>> 1. IET should be excluded from this list, iSCSI-SCST is IET updated
>> for SCST framework with a lot of bugfixes and improvements.
>>
>> 2. I think, everybody will agree that Linux iSCSI target should work
>> over some standard SCSI target framework. Hence the choice gets
>> narrower: SCST vs STGT. I don't think there's a way for a dedicated
>> iSCSI target (i.e. PyX/LIO) in the mainline, because of a lot of code
>> duplication. Nicholas could decide to move to either existing
>> framework (although, frankly, I don't think there's a possibility for
>> in-kernel iSCSI target and user space SCSI target framework) and if he
>> decide to go with SCST, I'll be glad to offer my help and support and
>> wouldn't care if LIO-SCST eventually replaced iSCSI-SCST. The better
>> one should win.
>
>
> why should linux as an iSCSI target be limited to passthrough to a SCSI
> device.
>
> the most common use of this sort of thing that I would see is to load up
> a bunch of 1TB SATA drives in a commodity PC, run software RAID, and
> then export the resulting volume to other servers via iSCSI. not a
> 'real' SCSI device in sight.
>
> As far as how good a standard iSCSI is, at this point I don't think it
> really matters. There are too many devices and manufacturers out there
> that implement iSCSI as their storage protocol (from both sides,
> offering storage to other systems, and using external storage).
> Sometimes the best technology doesn't win, but Linux should be
> interoperable with as much as possible and be ready to support the
> winners and the loosers in technology options, for as long as anyone
> chooses to use the old equipment (after all, we support things like
> Arcnet networking, which lost to Ethernet many years ago)
David, your question surprises me a lot. From where have you decided
that SCST supports only pass-through backstorage? Does the RAM disk,
which Bart has been using for performance tests, look like a SCSI device?
SCST supports all backstorage types you can imagine and Linux kernel
supports.
> David Lang
> -
> To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
> the body of a message to [email protected]
> More majordomo info at http://vger.kernel.org/majordomo-info.html
>
On Thu, 2008-02-07 at 14:51 -0800, [email protected] wrote:
> On Thu, 7 Feb 2008, Vladislav Bolkhovitin wrote:
>
> > Bart Van Assche wrote:
> >> - It has been discussed which iSCSI target implementation should be in
> >> the mainstream Linux kernel. There is no agreement on this subject
> >> yet. The short-term options are as follows:
> >> 1) Do not integrate any new iSCSI target implementation in the
> >> mainstream Linux kernel.
> >> 2) Add one of the existing in-kernel iSCSI target implementations to
> >> the kernel, e.g. SCST or PyX/LIO.
> >> 3) Create a new in-kernel iSCSI target implementation that combines
> >> the advantages of the existing iSCSI kernel target implementations
> >> (iETD, STGT, SCST and PyX/LIO).
> >>
> >> As an iSCSI user, I prefer option (3). The big question is whether the
> >> various storage target authors agree with this ?
> >
> > I tend to agree with some important notes:
> >
> > 1. IET should be excluded from this list, iSCSI-SCST is IET updated for SCST
> > framework with a lot of bugfixes and improvements.
> >
> > 2. I think, everybody will agree that Linux iSCSI target should work over
> > some standard SCSI target framework. Hence the choice gets narrower: SCST vs
> > STGT. I don't think there's a way for a dedicated iSCSI target (i.e. PyX/LIO)
> > in the mainline, because of a lot of code duplication. Nicholas could decide
> > to move to either existing framework (although, frankly, I don't think
> > there's a possibility for in-kernel iSCSI target and user space SCSI target
> > framework) and if he decide to go with SCST, I'll be glad to offer my help
> > and support and wouldn't care if LIO-SCST eventually replaced iSCSI-SCST. The
> > better one should win.
>
> why should linux as an iSCSI target be limited to passthrough to a SCSI
> device.
>
<nod>
I don't think anyone is saying it should be. It makes sense that the
more mature SCSI engines that have working code will be providing alot
of the foundation as we talk about options..
>From comparing the designs of SCST and LIO-SE, we know that SCST has
supports very SCSI specific target mode hardware, including software
target mode forks of other kernel code. This code for the target mode
pSCSI, FC and SAS control paths (more for the state machines, that CDB
emulation) that will most likely never need to be emulated on non SCSI
target engine. SCST has support for the most SCSI fabric protocols of
the group (although it is lacking iSER) while the LIO-SE only supports
traditional iSCSI using Linux/IP (this means TCP, SCTP and IPv6). The
design of LIO-SE was to make every iSCSI initiator that sends SCSI CDBs
and data to talk to every potential device in the Linux storage stack on
the largest amount of hardware architectures possible.
Most of the iSCSI Initiators I know (including non Linux) do not rely on
heavy SCSI task management, and I think this would be a lower priority
item to get real SCSI specific recovery in the traditional iSCSI target
for users. Espically things like SCSI target mode queue locking
(affectionally called Auto Contingent Allegiance) make no sense for
traditional iSCSI or iSER, because CmdSN rules are doing this for us.
> the most common use of this sort of thing that I would see is to load up a
> bunch of 1TB SATA drives in a commodity PC, run software RAID, and then
> export the resulting volume to other servers via iSCSI. not a 'real' SCSI
> device in sight.
>
I recently moved the last core LIO target machine from a hardware RAID5
to MD RAID6 with struct block_device exported LVM objects via
Linux/iSCSI to PVM and HVM domains, and I have been very happy with the
results. Being able to export any physical or virtual storage object
from whatever layer makes sense for your particular case. This applies
to both block and file level access. For example, making an iSCSI
Initiator and Target run in the most limited in environments places
where NAS (espically userspace server side) would have a really hard
time fitting, has always been a requirement. You can imagine a system
with a smaller amount of memory (say 32MB) having a difficult time doing
I/O to any amount of NAS clients.
If are talking about memory required to get best performance, using
kernel level DMA ring allocation and submission to a generic target
engine uses a significantly smaller amount of memory, than say
traditional buffered FILEIO. Going futher up the storage stack with
buffered file IO, regardless of if its block or file level, will always
start to add overhead. I think that kernel level FILEIO with O_DIRECT
and asyncio would probably help alot in this case for general target
mode usage of MD and LVM block devices.
This is because when we are using PSCSI or IBLOCK to queue I/Os which,
may need be different from the original IO from the initiator/client due
to OS storage subsystem differences and/or physical HBA limitiations for
the layers below block. The current LIO-SE API excepts the storage
object to present these physical limitiations if to engine they exist.
This is called iscsi_transport_t in iscsi_target_transport.h currently,
but really should be called something like target_subsytem_api_t and
plugins called target_pscsi_t, target_bio_t, target_file_t, etc.
> As far as how good a standard iSCSI is, at this point I don't think it
> really matters. There are too many devices and manufacturers out there
> that implement iSCSI as their storage protocol (from both sides, offering
> storage to other systems, and using external storage). Sometimes the best
> technology doesn't win, but Linux should be interoperable with as much as
> possible and be ready to support the winners and the loosers in technology
> options, for as long as anyone chooses to use the old equipment (after
> all, we support things like Arcnet networking, which lost to Ethernet many
> years ago)
>
The RFC-3720 standard has been stable for going on four years in 2008,
and as the implementations continue to mature, having Linux lead the way
in iSCSI Target, Initiator and Target/Initiator that can potentially run
on anything that can boot Linux on the many, many types of system and
storage around these days is the goal. I can't personally comment on
how many of these types of systems that target mode or iSCSI stacks have
run in other people's environments, but I have personally been involved
getting LIO/SE and Core/iSCSI running on i386 and x86_64, along with
Alpha, ia64, MIPS, PPC and POWER, and lots of ARM. I believe the LIO
Target and Initiator stacks have been able to run on the smallest
systems so far, including a uclinux 2.6 sub 100 Mhz board and ~4 MB of
usable sytem memory. This is still today with the LIO target stack,
which has been successfully run on the OpenMoko device with memory and
FILEIO. :-)
--nab
Btw, I definately agree that being able to export the large amount of
legacy drivers will continue to be an important part..
On Thu, 2008-02-07 at 12:37 -0800, Luben Tuikov wrote:
> Is there an open iSCSI Target implementation which does NOT
> issue commands to sub-target devices via the SCSI mid-layer, but
> bypasses it completely?
>
> Luben
>
Hi Luben,
I am guessing you mean futher down the stack, which I don't know this to
be the case. Going futher up the layers is the design of v2.9 LIO-SE.
There is a diagram explaining the basic concepts from a 10,000 foot
level.
http://linux-iscsi.org/builds/user/nab/storage-engine-concept.pdf
Note that only traditional iSCSI target is currently implemented in v2.9
LIO-SE codebase in the list of target mode fabrics on left side of the
layout. The API between the protocol headers that does
encoding/decoding target mode storage packets is probably the least
mature area of the LIO stack (because it has always been iSCSI looking
towards iSER :). I don't know who has the most mature API between the
storage engine and target storage protocol for doing this between SCST
and STGT, I am guessing SCST because of the difference in age of the
projects. Could someone be so kind to fill me in on this..?
Also note, the storage engine plugin for doing userspace passthrough on
the right is also currently not implemented. Userspace passthrough in
this context is an target engine I/O that is enforcing max_sector and
sector_size limitiations, and encodes/decodes target storage protocol
packets all out of view of userspace. The addressing will be completely
different if we are pointing SE target packets at non SCSI target ports
in userspace.
--nab
>
Nicholas A. Bellinger wrote:
>>>>- It has been discussed which iSCSI target implementation should be in
>>>>the mainstream Linux kernel. There is no agreement on this subject
>>>>yet. The short-term options are as follows:
>>>>1) Do not integrate any new iSCSI target implementation in the
>>>>mainstream Linux kernel.
>>>>2) Add one of the existing in-kernel iSCSI target implementations to
>>>>the kernel, e.g. SCST or PyX/LIO.
>>>>3) Create a new in-kernel iSCSI target implementation that combines
>>>>the advantages of the existing iSCSI kernel target implementations
>>>>(iETD, STGT, SCST and PyX/LIO).
>>>>
>>>>As an iSCSI user, I prefer option (3). The big question is whether the
>>>>various storage target authors agree with this ?
>>>
>>>I tend to agree with some important notes:
>>>
>>>1. IET should be excluded from this list, iSCSI-SCST is IET updated for SCST
>>>framework with a lot of bugfixes and improvements.
>>>
>>>2. I think, everybody will agree that Linux iSCSI target should work over
>>>some standard SCSI target framework. Hence the choice gets narrower: SCST vs
>>>STGT. I don't think there's a way for a dedicated iSCSI target (i.e. PyX/LIO)
>>>in the mainline, because of a lot of code duplication. Nicholas could decide
>>>to move to either existing framework (although, frankly, I don't think
>>>there's a possibility for in-kernel iSCSI target and user space SCSI target
>>>framework) and if he decide to go with SCST, I'll be glad to offer my help
>>>and support and wouldn't care if LIO-SCST eventually replaced iSCSI-SCST. The
>>>better one should win.
>>
>>why should linux as an iSCSI target be limited to passthrough to a SCSI
>>device.
>
> <nod>
>
> I don't think anyone is saying it should be. It makes sense that the
> more mature SCSI engines that have working code will be providing alot
> of the foundation as we talk about options..
>
>>From comparing the designs of SCST and LIO-SE, we know that SCST has
> supports very SCSI specific target mode hardware, including software
> target mode forks of other kernel code. This code for the target mode
> pSCSI, FC and SAS control paths (more for the state machines, that CDB
> emulation) that will most likely never need to be emulated on non SCSI
> target engine.
...but required for SCSI. So, it must be, anyway.
> SCST has support for the most SCSI fabric protocols of
> the group (although it is lacking iSER) while the LIO-SE only supports
> traditional iSCSI using Linux/IP (this means TCP, SCTP and IPv6). The
> design of LIO-SE was to make every iSCSI initiator that sends SCSI CDBs
> and data to talk to every potential device in the Linux storage stack on
> the largest amount of hardware architectures possible.
>
> Most of the iSCSI Initiators I know (including non Linux) do not rely on
> heavy SCSI task management, and I think this would be a lower priority
> item to get real SCSI specific recovery in the traditional iSCSI target
> for users. Espically things like SCSI target mode queue locking
> (affectionally called Auto Contingent Allegiance) make no sense for
> traditional iSCSI or iSER, because CmdSN rules are doing this for us.
Sorry, it isn't correct. ACA provides possibility to lock commands queue
in case of CHECK CONDITION, so allows to keep commands execution order
in case of errors. CmdSN keeps commands execution order only in case of
success, in case of error the next queued command will be executed
immediately after the failed one, although application might require to
have all subsequent after the failed one commands aborted. Think about
journaled file systems, for instance. Also ACA allows to retry the
failed command and then resume the queue.
Vlad
Nicholas A. Bellinger wrote:
> On Thu, 2008-02-07 at 12:37 -0800, Luben Tuikov wrote:
>
>>Is there an open iSCSI Target implementation which does NOT
>>issue commands to sub-target devices via the SCSI mid-layer, but
>>bypasses it completely?
>>
>> Luben
>>
>
>
> Hi Luben,
>
> I am guessing you mean futher down the stack, which I don't know this to
> be the case. Going futher up the layers is the design of v2.9 LIO-SE.
> There is a diagram explaining the basic concepts from a 10,000 foot
> level.
>
> http://linux-iscsi.org/builds/user/nab/storage-engine-concept.pdf
>
> Note that only traditional iSCSI target is currently implemented in v2.9
> LIO-SE codebase in the list of target mode fabrics on left side of the
> layout. The API between the protocol headers that does
> encoding/decoding target mode storage packets is probably the least
> mature area of the LIO stack (because it has always been iSCSI looking
> towards iSER :). I don't know who has the most mature API between the
> storage engine and target storage protocol for doing this between SCST
> and STGT, I am guessing SCST because of the difference in age of the
> projects. Could someone be so kind to fill me in on this..?
SCST uses scsi_execute_async_fifo() function to submit commands to SCSI
devices in the pass-through mode. This function is slightly modified
version of scsi_execute_async(), which submits requests in FIFO order
instead of LIFO as scsi_execute_async() does (so with
scsi_execute_async() they are executed in the reverse order).
Scsi_execute_async_fifo() added as a separate patch to the kernel.
> Also note, the storage engine plugin for doing userspace passthrough on
> the right is also currently not implemented. Userspace passthrough in
> this context is an target engine I/O that is enforcing max_sector and
> sector_size limitiations, and encodes/decodes target storage protocol
> packets all out of view of userspace. The addressing will be completely
> different if we are pointing SE target packets at non SCSI target ports
> in userspace.
>
> --nab
On Fri, 2008-02-08 at 17:36 +0300, Vladislav Bolkhovitin wrote:
> Nicholas A. Bellinger wrote:
> >>>>- It has been discussed which iSCSI target implementation should be in
> >>>>the mainstream Linux kernel. There is no agreement on this subject
> >>>>yet. The short-term options are as follows:
> >>>>1) Do not integrate any new iSCSI target implementation in the
> >>>>mainstream Linux kernel.
> >>>>2) Add one of the existing in-kernel iSCSI target implementations to
> >>>>the kernel, e.g. SCST or PyX/LIO.
> >>>>3) Create a new in-kernel iSCSI target implementation that combines
> >>>>the advantages of the existing iSCSI kernel target implementations
> >>>>(iETD, STGT, SCST and PyX/LIO).
> >>>>
> >>>>As an iSCSI user, I prefer option (3). The big question is whether the
> >>>>various storage target authors agree with this ?
> >>>
> >>>I tend to agree with some important notes:
> >>>
> >>>1. IET should be excluded from this list, iSCSI-SCST is IET updated for SCST
> >>>framework with a lot of bugfixes and improvements.
> >>>
> >>>2. I think, everybody will agree that Linux iSCSI target should work over
> >>>some standard SCSI target framework. Hence the choice gets narrower: SCST vs
> >>>STGT. I don't think there's a way for a dedicated iSCSI target (i.e. PyX/LIO)
> >>>in the mainline, because of a lot of code duplication. Nicholas could decide
> >>>to move to either existing framework (although, frankly, I don't think
> >>>there's a possibility for in-kernel iSCSI target and user space SCSI target
> >>>framework) and if he decide to go with SCST, I'll be glad to offer my help
> >>>and support and wouldn't care if LIO-SCST eventually replaced iSCSI-SCST. The
> >>>better one should win.
> >>
> >>why should linux as an iSCSI target be limited to passthrough to a SCSI
> >>device.
> >
> > <nod>
> >
> > I don't think anyone is saying it should be. It makes sense that the
> > more mature SCSI engines that have working code will be providing alot
> > of the foundation as we talk about options..
> >
> >>From comparing the designs of SCST and LIO-SE, we know that SCST has
> > supports very SCSI specific target mode hardware, including software
> > target mode forks of other kernel code. This code for the target mode
> > pSCSI, FC and SAS control paths (more for the state machines, that CDB
> > emulation) that will most likely never need to be emulated on non SCSI
> > target engine.
>
> ...but required for SCSI. So, it must be, anyway.
>
> > SCST has support for the most SCSI fabric protocols of
> > the group (although it is lacking iSER) while the LIO-SE only supports
> > traditional iSCSI using Linux/IP (this means TCP, SCTP and IPv6). The
> > design of LIO-SE was to make every iSCSI initiator that sends SCSI CDBs
> > and data to talk to every potential device in the Linux storage stack on
> > the largest amount of hardware architectures possible.
> >
> > Most of the iSCSI Initiators I know (including non Linux) do not rely on
> > heavy SCSI task management, and I think this would be a lower priority
> > item to get real SCSI specific recovery in the traditional iSCSI target
> > for users. Espically things like SCSI target mode queue locking
> > (affectionally called Auto Contingent Allegiance) make no sense for
> > traditional iSCSI or iSER, because CmdSN rules are doing this for us.
>
> Sorry, it isn't correct. ACA provides possibility to lock commands queue
> in case of CHECK CONDITION, so allows to keep commands execution order
> in case of errors. CmdSN keeps commands execution order only in case of
> success, in case of error the next queued command will be executed
> immediately after the failed one, although application might require to
> have all subsequent after the failed one commands aborted. Think about
> journaled file systems, for instance. Also ACA allows to retry the
> failed command and then resume the queue.
>
Fair enough. The point I was making is that I have never actually seen
an iSCSI Initiator use ACA functionality (I don't believe that the Linux
SCSI Ml implements this), or actually generate a CLEAR_ACA task
management request.
--nab
> Vlad
>
On Fri, 2008-02-08 at 17:42 +0300, Vladislav Bolkhovitin wrote:
> Nicholas A. Bellinger wrote:
> > On Thu, 2008-02-07 at 12:37 -0800, Luben Tuikov wrote:
> >
> >>Is there an open iSCSI Target implementation which does NOT
> >>issue commands to sub-target devices via the SCSI mid-layer, but
> >>bypasses it completely?
> >>
> >> Luben
> >>
> >
> >
> > Hi Luben,
> >
> > I am guessing you mean futher down the stack, which I don't know this to
> > be the case. Going futher up the layers is the design of v2.9 LIO-SE.
> > There is a diagram explaining the basic concepts from a 10,000 foot
> > level.
> >
> > http://linux-iscsi.org/builds/user/nab/storage-engine-concept.pdf
> >
> > Note that only traditional iSCSI target is currently implemented in v2.9
> > LIO-SE codebase in the list of target mode fabrics on left side of the
> > layout. The API between the protocol headers that does
> > encoding/decoding target mode storage packets is probably the least
> > mature area of the LIO stack (because it has always been iSCSI looking
> > towards iSER :). I don't know who has the most mature API between the
> > storage engine and target storage protocol for doing this between SCST
> > and STGT, I am guessing SCST because of the difference in age of the
> > projects. Could someone be so kind to fill me in on this..?
>
> SCST uses scsi_execute_async_fifo() function to submit commands to SCSI
> devices in the pass-through mode. This function is slightly modified
> version of scsi_execute_async(), which submits requests in FIFO order
> instead of LIFO as scsi_execute_async() does (so with
> scsi_execute_async() they are executed in the reverse order).
> Scsi_execute_async_fifo() added as a separate patch to the kernel.
The LIO-SE PSCSI Plugin also depends on scsi_execute_async() for builds
on >= 2.6.18. Note in the core LIO storage engine code (would be
iscsi_target_transport.c), there is no subsystem dependence logic. The
LIO-SE API is what allows the SE plugins to remain simple and small:
-rw-r--r-- 1 root root 35008 2008-02-02 03:25 iscsi_target_pscsi.c
-rw-r--r-- 1 root root 7537 2008-02-02 17:27 iscsi_target_pscsi.h
-rw-r--r-- 1 root root 18269 2008-02-04 02:23 iscsi_target_iblock.c
-rw-r--r-- 1 root root 6834 2008-02-04 02:25 iscsi_target_iblock.h
-rw-r--r-- 1 root root 30611 2008-02-02 03:25 iscsi_target_file.c
-rw-r--r-- 1 root root 7833 2008-02-02 17:27 iscsi_target_file.h
-rw-r--r-- 1 root root 35154 2008-02-02 04:01 iscsi_target_rd.c
-rw-r--r-- 1 root root 9900 2008-02-02 17:27 iscsi_target_rd.h
It also means that the core LIO-SE code does not have to change when the
subsystem APIs change. This has been important in the past for the
project, but for upstream code, probably would not make a huge
difference.
--nab
On Fri, 8 Feb 2008, Vladislav Bolkhovitin wrote:
>>> 2. I think, everybody will agree that Linux iSCSI target should work over
>>> some standard SCSI target framework. Hence the choice gets narrower: SCST
>>> vs STGT. I don't think there's a way for a dedicated iSCSI target (i.e.
>>> PyX/LIO) in the mainline, because of a lot of code duplication. Nicholas
>>> could decide to move to either existing framework (although, frankly, I
>>> don't think there's a possibility for in-kernel iSCSI target and user
>>> space SCSI target framework) and if he decide to go with SCST, I'll be
>>> glad to offer my help and support and wouldn't care if LIO-SCST eventually
>>> replaced iSCSI-SCST. The better one should win.
>>
>>
>> why should linux as an iSCSI target be limited to passthrough to a SCSI
>> device.
>>
>> the most common use of this sort of thing that I would see is to load up a
>> bunch of 1TB SATA drives in a commodity PC, run software RAID, and then
>> export the resulting volume to other servers via iSCSI. not a 'real' SCSI
>> device in sight.
>>
> David, your question surprises me a lot. From where have you decided that
> SCST supports only pass-through backstorage? Does the RAM disk, which Bart
> has been using for performance tests, look like a SCSI device?
I was responding to the start of item #2 that I left in the quote above.
it asn't saying that SCST didn't support that, but was stating that any
implementation of a iSCSI target should use the SCSI framework. I read
this to mean that this would only be able to access things that the SCSI
framework can access, and that would not be things like ramdisks, raid
arrays, etc.
David Lang
--- On Fri, 2/8/08, Vladislav Bolkhovitin <[email protected]> wrote:
> > Is there an open iSCSI Target implementation which
> does NOT
> > issue commands to sub-target devices via the SCSI
> mid-layer, but
> > bypasses it completely?
>
> What do you mean? To call directly low level backstorage
> SCSI drivers
> queuecommand() routine? What are advantages of it?
Yes, that's what I meant. Just curious.
Thanks,
Luben
Luben Tuikov wrote:
>>>Is there an open iSCSI Target implementation which
>>
>>does NOT
>>
>>>issue commands to sub-target devices via the SCSI
>>
>>mid-layer, but
>>
>>>bypasses it completely?
>>
>>What do you mean? To call directly low level backstorage
>>SCSI drivers
>>queuecommand() routine? What are advantages of it?
>
> Yes, that's what I meant. Just curious.
What's advantage of it?
> Thanks,
> Luben
On Feb 6, 2008 1:11 AM, Nicholas A. Bellinger <[email protected]> wrote:
> I have always observed the case with LIO SE/iSCSI target mode ...
Hello Nicholas,
Are you sure that the LIO-SE kernel module source code is ready for
inclusion in the mainstream Linux kernel ? As you know I tried to test
the LIO-SE iSCSI target. Already while configuring the target I
encountered a kernel crash that froze the whole system. I can
reproduce this kernel crash easily, and I reported it 11 days ago on
the LIO-SE mailing list (February 4, 2008). One of the call stacks I
posted shows a crash in mempool_alloc() called from jbd. Or: the crash
is most likely the result of memory corruption caused by LIO-SE.
Because I was curious to know why it took so long to fix such a severe
crash, I started browsing through the LIO-SE source code. Analysis of
the LIO-SE kernel module source code learned me that this crash is not
a coincidence. Dynamic memory allocation (kmalloc()/kfree()) in the
LIO-SE kernel module is complex and hard to verify. There are 412
memory allocation/deallocation calls in the current version of the
LIO-SE kernel module source code, which is a lot. Additionally,
because of the complexity of the memory handling in LIO-SE, it is not
possible to verify the correctness of the memory handling by analyzing
a single function at a time. In my opinion this makes the LIO-SE
source code hard to maintain.
Furthermore, the LIO-SE kernel module source code does not follow
conventions that have proven their value in the past like grouping all
error handling at the end of a function. As could be expected, the
consequence is that error handling is not correct in several
functions, resulting in memory leaks in case of an error. Some
examples of functions in which error handling is clearly incorrect:
* transport_allocate_passthrough().
* iscsi_do_build_list().
Bart Van Assche.
Greetings all,
On Tue, 2008-02-12 at 17:05 +0100, Bart Van Assche wrote:
> On Feb 6, 2008 1:11 AM, Nicholas A. Bellinger <[email protected]> wrote:
> > I have always observed the case with LIO SE/iSCSI target mode ...
>
> Hello Nicholas,
>
> Are you sure that the LIO-SE kernel module source code is ready for
> inclusion in the mainstream Linux kernel ? As you know I tried to test
> the LIO-SE iSCSI target. Already while configuring the target I
> encountered a kernel crash that froze the whole system. I can
> reproduce this kernel crash easily, and I reported it 11 days ago on
> the LIO-SE mailing list (February 4, 2008). One of the call stacks I
> posted shows a crash in mempool_alloc() called from jbd. Or: the crash
> is most likely the result of memory corruption caused by LIO-SE.
>
So I was able to FINALLY track this down to:
-# CONFIG_SLUB_DEBUG is not set
-# CONFIG_SLAB is not set
-CONFIG_SLUB=y
+CONFIG_SLAB=y
in both your and Chris Weiss's configs that was causing the
reproduceable general protection faults. I also disabled
CONFIG_RELOCATABLE and crash dump because I was debugging using kdb in
x86_64 VM on 2.6.24 with your config. I am pretty sure you can leave
this (crash dump) in your config for testing.
This can take a while to compile and take up alot of space, esp. with
all of the kernel debug options enabled, which on 2.6.24, really amounts
to alot of CPU time when building. Also with your original config, I
was seeing some strange undefined module objects after Stage 2 Link with
iscsi_target_mod with modpost with the SLUB the lockups (which are not
random btw, and are tracked back to __kmalloc()).. Also, at module load
time with the original config, there where some warning about symbol
objects (I believe it was SCSI related, same as the ones with modpost).
In any event, the dozen 1000 loop discovery test is now working fine (as
well as IPoIB) with the above config change, and you should be ready to
go for your testing.
Tomo, Vlad, Andrew and Co:
Do you have any ideas why this would be the case with LIO-Target..? Is
anyone else seeing something similar to this with their target mode
(mabye its all out of tree code..?) that is having an issue..? I am
using Debian x86_64 and Bart and Chris are using Ubuntu x86_64 and we
both have this problem with CONFIG_SLUB on >= 2.6.22 kernel.org
kernels.
Also, I will recompile some of my non x86 machines with the above
enabled and see if I can reproduce.. Here the Bart's config again:
http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/30835aede1028188
> Because I was curious to know why it took so long to fix such a severe
> crash, I started browsing through the LIO-SE source code. Analysis of
> the LIO-SE kernel module source code learned me that this crash is not
> a coincidence. Dynamic memory allocation (kmalloc()/kfree()) in the
> LIO-SE kernel module is complex and hard to verify.
What the LIO-SE Target module does is complex. :P Sorry for taking so
long, I had to start tracking this down by CONFIG_ option with your
config on an x86_64 VM.
> There are 412
> memory allocation/deallocation calls in the current version of the
> LIO-SE kernel module source code, which is a lot. Additionally,
> because of the complexity of the memory handling in LIO-SE, it is not
> possible to verify the correctness of the memory handling by analyzing
> a single function at a time. In my opinion this makes the LIO-SE
> source code hard to maintain.
> Furthermore, the LIO-SE kernel module source code does not follow
> conventions that have proven their value in the past like grouping all
> error handling at the end of a function. As could be expected, the
> consequence is that error handling is not correct in several
> functions, resulting in memory leaks in case of an error.
I would be more than happy to point the release paths for iSCSI Target
and LIO-SE to show they are not actual memory leaks (as I mentioned,
this code has been stable for a number of years) for some particular SE
or iSCSI Target logic if you are interested..
Also, if we are talking about target mode storage engine that should be
going upstream, the API to the current stable and future storage
systems, and of course the Mem->SG and SG->Mem that handles all possible
cases of max_sectors and sector_size to past, present, and future. I
really glad that you have been taking a look at this, because some of
the code (as you mention) can get very complex to make this a reality as
it has been with LIO-Target since v2.2.
> Some
> examples of functions in which error handling is clearly incorrect:
> * transport_allocate_passthrough().
> * iscsi_do_build_list().
>
You did find the one in transport_allocate_passthrough() and the
strncpy() + strlen() in userspace. Also, thanks for pointing me to the
missing sg_init_table() and sg_mark_end() usage for 2.6.24. I will post
an update to my thread about how to do this for other drivers..
I will have a look at your new changes and post them on LIO-Target-Dev
for your review. Please feel free to Ack them when I post.
(Thanks Bart !!)
PS: Sometimes it takes a while when you are on the bleeding edge of
development to track these types of issues down. :-)
--nab
On Tue, 2008-02-12 at 19:57 -0800, Nicholas A. Bellinger wrote:
> Greetings all,
>
> On Tue, 2008-02-12 at 17:05 +0100, Bart Van Assche wrote:
> > On Feb 6, 2008 1:11 AM, Nicholas A. Bellinger <[email protected]> wrote:
> > > I have always observed the case with LIO SE/iSCSI target mode ...
> >
> > Hello Nicholas,
> >
> > Are you sure that the LIO-SE kernel module source code is ready for
> > inclusion in the mainstream Linux kernel ? As you know I tried to test
> > the LIO-SE iSCSI target. Already while configuring the target I
> > encountered a kernel crash that froze the whole system. I can
> > reproduce this kernel crash easily, and I reported it 11 days ago on
> > the LIO-SE mailing list (February 4, 2008). One of the call stacks I
> > posted shows a crash in mempool_alloc() called from jbd. Or: the crash
> > is most likely the result of memory corruption caused by LIO-SE.
> >
>
> So I was able to FINALLY track this down to:
>
> -# CONFIG_SLUB_DEBUG is not set
> -# CONFIG_SLAB is not set
> -CONFIG_SLUB=y
> +CONFIG_SLAB=y
>
> in both your and Chris Weiss's configs that was causing the
> reproduceable general protection faults. I also disabled
> CONFIG_RELOCATABLE and crash dump because I was debugging using kdb in
> x86_64 VM on 2.6.24 with your config. I am pretty sure you can leave
> this (crash dump) in your config for testing.
>
> This can take a while to compile and take up alot of space, esp. with
> all of the kernel debug options enabled, which on 2.6.24, really amounts
> to alot of CPU time when building. Also with your original config, I
> was seeing some strange undefined module objects after Stage 2 Link with
> iscsi_target_mod with modpost with the SLUB the lockups (which are not
> random btw, and are tracked back to __kmalloc()).. Also, at module load
> time with the original config, there where some warning about symbol
> objects (I believe it was SCSI related, same as the ones with modpost).
>
> In any event, the dozen 1000 loop discovery test is now working fine (as
> well as IPoIB) with the above config change, and you should be ready to
> go for your testing.
>
> Tomo, Vlad, Andrew and Co:
>
> Do you have any ideas why this would be the case with LIO-Target..? Is
> anyone else seeing something similar to this with their target mode
> (mabye its all out of tree code..?) that is having an issue..? I am
> using Debian x86_64 and Bart and Chris are using Ubuntu x86_64 and we
> both have this problem with CONFIG_SLUB on >= 2.6.22 kernel.org
> kernels.
>
> Also, I will recompile some of my non x86 machines with the above
> enabled and see if I can reproduce.. Here the Bart's config again:
>
> http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/30835aede1028188
>
This is also failing on CONFIG_SLUB on 2.6.24 ppc64. Since the rest of
the system seems to work fine, my only guess it may be related to the
fact that the module is being compiled out of tree. I took a quick
glance at what kbuild was using for compiler and linker parameters, but
nothing looked out of the ordinary.
I will take a look with kdb and SLUB re-enabled on x86_64 and see if this
helps shed any light on the issue. Is anyone else seeing an issue with CONFIG_SLUB..?
Also, I wonder who else aside from Ubuntu is using this by default in
their .config..?
--nab
On Tue, 2008-02-12 at 22:18 -0800, Nicholas A. Bellinger wrote:
> On Tue, 2008-02-12 at 19:57 -0800, Nicholas A. Bellinger wrote:
> > Greetings all,
> >
> > On Tue, 2008-02-12 at 17:05 +0100, Bart Van Assche wrote:
> > > On Feb 6, 2008 1:11 AM, Nicholas A. Bellinger <[email protected]> wrote:
> > > > I have always observed the case with LIO SE/iSCSI target mode ...
> > >
> > > Hello Nicholas,
> > >
> > > Are you sure that the LIO-SE kernel module source code is ready for
> > > inclusion in the mainstream Linux kernel ? As you know I tried to test
> > > the LIO-SE iSCSI target. Already while configuring the target I
> > > encountered a kernel crash that froze the whole system. I can
> > > reproduce this kernel crash easily, and I reported it 11 days ago on
> > > the LIO-SE mailing list (February 4, 2008). One of the call stacks I
> > > posted shows a crash in mempool_alloc() called from jbd. Or: the crash
> > > is most likely the result of memory corruption caused by LIO-SE.
> > >
> >
> > So I was able to FINALLY track this down to:
> >
> > -# CONFIG_SLUB_DEBUG is not set
> > -# CONFIG_SLAB is not set
> > -CONFIG_SLUB=y
> > +CONFIG_SLAB=y
> >
> > in both your and Chris Weiss's configs that was causing the
> > reproduceable general protection faults. I also disabled
> > CONFIG_RELOCATABLE and crash dump because I was debugging using kdb in
> > x86_64 VM on 2.6.24 with your config. I am pretty sure you can leave
> > this (crash dump) in your config for testing.
> >
> > This can take a while to compile and take up alot of space, esp. with
> > all of the kernel debug options enabled, which on 2.6.24, really amounts
> > to alot of CPU time when building. Also with your original config, I
> > was seeing some strange undefined module objects after Stage 2 Link with
> > iscsi_target_mod with modpost with the SLUB the lockups (which are not
> > random btw, and are tracked back to __kmalloc()).. Also, at module load
> > time with the original config, there where some warning about symbol
> > objects (I believe it was SCSI related, same as the ones with modpost).
> >
> > In any event, the dozen 1000 loop discovery test is now working fine (as
> > well as IPoIB) with the above config change, and you should be ready to
> > go for your testing.
> >
> > Tomo, Vlad, Andrew and Co:
> >
> > Do you have any ideas why this would be the case with LIO-Target..? Is
> > anyone else seeing something similar to this with their target mode
> > (mabye its all out of tree code..?) that is having an issue..? I am
> > using Debian x86_64 and Bart and Chris are using Ubuntu x86_64 and we
> > both have this problem with CONFIG_SLUB on >= 2.6.22 kernel.org
> > kernels.
> >
> > Also, I will recompile some of my non x86 machines with the above
> > enabled and see if I can reproduce.. Here the Bart's config again:
> >
> > http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/30835aede1028188
> >
>
> This is also failing on CONFIG_SLUB on 2.6.24 ppc64. Since the rest of
> the system seems to work fine, my only guess it may be related to the
> fact that the module is being compiled out of tree. I took a quick
> glance at what kbuild was using for compiler and linker parameters, but
> nothing looked out of the ordinary.
>
> I will take a look with kdb and SLUB re-enabled on x86_64 and see if this
> helps shed any light on the issue. Is anyone else seeing an issue with CONFIG_SLUB..?
>
I was able to track this down to a memory corruption issue in the
in-band iSCSI discovery path. I just made the change, and the diff can
be located at:
http://groups.google.com/group/linux-iscsi-target-dev/browse_thread/thread/a70d4835c55be392
In any event, this is now fixed, and I will be generating some new
builds for LIO-Target shortly for Debian, CentOS and Ubuntu. Espically
for the Ubuntu folks, where this is going to be an issue with their
default kernel config.
A big thanks to Bart Van Assche for helping me locate the actual issue,
and giving me a clue with slub_debug=FZPU. Thanks again,
--nab
On Thu, Feb 7, 2008 at 2:45 PM, Vladislav Bolkhovitin <[email protected]> wrote:
>
> Bart Van Assche wrote:
> > Since the focus of this thread shifted somewhat in the last few
> > messages, I'll try to summarize what has been discussed so far:
> > - There was a number of participants who joined this discussion
> > spontaneously. This suggests that there is considerable interest in
> > networked storage and iSCSI.
> > - It has been motivated why iSCSI makes sense as a storage protocol
> > (compared to ATA over Ethernet and Fibre Channel over Ethernet).
> > - The direct I/O performance results for block transfer sizes below 64
> > KB are a meaningful benchmark for storage target implementations.
> > - It has been discussed whether an iSCSI target should be implemented
> > in user space or in kernel space. It is clear now that an
> > implementation in the kernel can be made faster than a user space
> > implementation (http://kerneltrap.org/mailarchive/linux-kernel/2008/2/4/714804).
> > Regarding existing implementations, measurements have a.o. shown that
> > SCST is faster than STGT (30% with the following setup: iSCSI via
> > IPoIB and direct I/O block transfers with a size of 512 bytes).
> > - It has been discussed which iSCSI target implementation should be in
> > the mainstream Linux kernel. There is no agreement on this subject
> > yet. The short-term options are as follows:
> > 1) Do not integrate any new iSCSI target implementation in the
> > mainstream Linux kernel.
> > 2) Add one of the existing in-kernel iSCSI target implementations to
> > the kernel, e.g. SCST or PyX/LIO.
> > 3) Create a new in-kernel iSCSI target implementation that combines
> > the advantages of the existing iSCSI kernel target implementations
> > (iETD, STGT, SCST and PyX/LIO).
> >
> > As an iSCSI user, I prefer option (3). The big question is whether the
> > various storage target authors agree with this ?
>
> I tend to agree with some important notes:
>
> 1. IET should be excluded from this list, iSCSI-SCST is IET updated for
> SCST framework with a lot of bugfixes and improvements.
>
> 2. I think, everybody will agree that Linux iSCSI target should work
> over some standard SCSI target framework. Hence the choice gets
> narrower: SCST vs STGT. I don't think there's a way for a dedicated
> iSCSI target (i.e. PyX/LIO) in the mainline, because of a lot of code
> duplication. Nicholas could decide to move to either existing framework
> (although, frankly, I don't think there's a possibility for in-kernel
> iSCSI target and user space SCSI target framework) and if he decide to
> go with SCST, I'll be glad to offer my help and support and wouldn't
> care if LIO-SCST eventually replaced iSCSI-SCST. The better one should win.
If I understood the above correctly, regarding a kernel space iSCSI
target implementation, only LIO-SE and SCST should be considered. What
I know today about these Linux iSCSI target implementations is as
follows:
* SCST performs slightly better than LIO-SE, and LIO-SE performs
slightly better than STGT (both with regard to latency and with regard
to bandwidth).
* The coding style of SCST is closer to the Linux kernel coding style
than the coding style of the LIO-SE project.
* The structure of SCST is closer to what Linus expects than the
structure of LIO-SE (i.e., authentication handled in userspace, data
transfer handled by the kernel -- LIO-SE handles both in kernel
space).
* Until now I did not encounter any strange behavior in SCST. The
issues I encountered with LIO-SE are being resolved via the LIO-SE
mailing list (http://groups.google.com/group/linux-iscsi-target-dev).
It would take too much effort to develop a new kernel space iSCSI
target from scratch -- we should start from either LIO-SE or SCST. My
opinion is that the best approach is to start with integrating SCST in
the mainstream kernel, and that the more advanced features from LIO-SE
that are not yet in SCST can be ported from LIO-SE to the SCST
framework.
Nicholas, do you think the structure of SCST is powerful enough to be
extended with LIO-SE's powerful features like ERL-2 ?
Bart Van Assche.
Bart Van Assche wrote:
> On Feb 5, 2008 6:01 PM, Erez Zilber <[email protected]> wrote:
>
>> Using such large values for FirstBurstLength will give you poor
>> performance numbers for WRITE commands (with iSER). FirstBurstLength
>> means how much data should you send as unsolicited data (i.e. without
>> RDMA). It means that your WRITE commands were sent without RDMA.
>>
>
> Sorry, but I'm afraid you got this wrong. When the iSER transport is
> used instead of TCP, all data is sent via RDMA, including unsolicited
> data. If you have look at the iSER implementation in the Linux kernel
> (source files under drivers/infiniband/ulp/iser), you will see that
> all data is transferred via RDMA and not via TCP/IP.
>
>
When you execute WRITE commands with iSCSI, it works like this:
EDTL (Expected data length) - the data length of your command
FirstBurstLength - the length of data that will be sent as unsolicited
data (i.e. as immediate data with the SCSI command and as unsolicited
data-out PDUs)
If you use a high value for FirstBurstLength, all (or most) of your data
will be sent as unsolicited data-out PDUs. These PDUs don't use the RDMA
engine, so you miss the advantage of IB.
If you use a lower value for FirstBurstLength, EDTL - FirstBurstLength
bytes will be sent as solicited data-out PDUs. With iSER, solicited
data-out PDUs are RDMA operations.
I hope that I'm more clear now.
Erez
On Feb 18, 2008 10:43 AM, Erez Zilber <[email protected]> wrote:
> If you use a high value for FirstBurstLength, all (or most) of your data
> will be sent as unsolicited data-out PDUs. These PDUs don't use the RDMA
> engine, so you miss the advantage of IB.
Hello Erez,
Did you notice the e-mail Roland Dreier wrote on Februari 6, 2008 ?
This is what Roland wrote:
> I think the confusion here is caused by a slight misuse of the term
> "RDMA". It is true that all data is always transported over an
> InfiniBand connection when iSER is used, but not all such transfers
> are one-sided RDMA operations; some data can be transferred using
> send/receive operations.
Or: data sent during the first burst is not transferred via one-sided
remote memory reads or writes but via two-sided send/receive
operations. At least on my setup, these operations are as fast as
one-sided remote memory reads or writes. As an example, I obtained the
following numbers on my setup (SDR 4x network);
ib_write_bw: 933 MB/s.
ib_read_bw: 905 MB/s.
ib_send_bw: 931 MB/s.
Bart Van Assche.
Bart Van Assche wrote:
> On Feb 18, 2008 10:43 AM, Erez Zilber <[email protected]> wrote:
>
>> If you use a high value for FirstBurstLength, all (or most) of your data
>> will be sent as unsolicited data-out PDUs. These PDUs don't use the RDMA
>> engine, so you miss the advantage of IB.
>>
>
> Hello Erez,
>
> Did you notice the e-mail Roland Dreier wrote on Februari 6, 2008 ?
> This is what Roland wrote:
>
>> I think the confusion here is caused by a slight misuse of the term
>> "RDMA". It is true that all data is always transported over an
>> InfiniBand connection when iSER is used, but not all such transfers
>> are one-sided RDMA operations; some data can be transferred using
>> send/receive operations.
>>
>
>
Yes, I saw that. I tried to give an explanation with more details.
> Or: data sent during the first burst is not transferred via one-sided
> remote memory reads or writes but via two-sided send/receive
> operations. At least on my setup, these operations are as fast as
> one-sided remote memory reads or writes. As an example, I obtained the
> following numbers on my setup (SDR 4x network);
> ib_write_bw: 933 MB/s.
> ib_read_bw: 905 MB/s.
> ib_send_bw: 931 MB/s.
>
>
According to these numbers one can think that you don't need RDMA at
all, just send iSCSI PDUs over IB. The benchmarks that you use are
synthetic IB benchmarks that are not equivalent to iSCSI over iSER. They
just send IB packets. I'm not surprised that you got more or less the
same performance because, AFAIK, ib_send_bw doesn't copy data (unlike
iSCSI that has to copy data that is sent/received without RDMA).
When you use RDMA with iSCSI (i.e. iSER), you don't need to create iSCSI
PDUs and process them. The CPU is not busy as it is with iSCSI over TCP
because no data copies are required. Another advantage is that you don't
need header/data digest because the IB HW does that.
Erez
On Feb 20, 2008 8:34 AM, Erez Zilber <[email protected]> wrote:
> Bart Van Assche wrote:
> > Or: data sent during the first burst is not transferred via one-sided
> > remote memory reads or writes but via two-sided send/receive
> > operations. At least on my setup, these operations are as fast as
> > one-sided remote memory reads or writes. As an example, I obtained the
> > following numbers on my setup (SDR 4x network);
> > ib_write_bw: 933 MB/s.
> > ib_read_bw: 905 MB/s.
> > ib_send_bw: 931 MB/s.
>
> According to these numbers one can think that you don't need RDMA at
> all, just send iSCSI PDUs over IB.
Sorry, but you are misinterpreting what I wrote.
> The benchmarks that you use are
> synthetic IB benchmarks that are not equivalent to iSCSI over iSER. They
> just send IB packets. I'm not surprised that you got more or less the
> same performance because, AFAIK, ib_send_bw doesn't copy data (unlike
> iSCSI that has to copy data that is sent/received without RDMA).
I agree that ib_write_bw / ib_read_bw / ib_send_bw performance results
are not equivalent to iSCSI over iSER. The reason that I included
these performance results was to illustrate that two-sided data
transfers over IB are about as fast as one-sided data transfers.
> When you use RDMA with iSCSI (i.e. iSER), you don't need to create iSCSI
> PDUs and process them. The CPU is not busy as it is with iSCSI over TCP
> because no data copies are required. Another advantage is that you don't
> need header/data digest because the IB HW does that.
As far as I know, when using iSER, the FirstBurstLength bytes of data
are sent via two-sided data transfers, and there is no CPU
intervention required to transfer the data itself over the IB network.
Bart Van Assche.