I'm re-issuing Hannes's latency patches in preparation for LSFMM
Changes since V2:
I've done quite a bit of work cleaning up these patches. There were a
number of checkpatch.pl problems as well as some compile time errors
when config BLK_NODE_LATENCY was turned off. After the clean up I
rebased these patches onto Ewan's "nvme: queue-depth multipath iopolicy"
patches. This allowed me to test both iopolicy changes together.
All of my test results, together with the scripts I used to generate these
graphs, are available at:
https://github.com/johnmeneghini/iopolicy
Please use the scripts in this repository to do your own testing.
Changes since V1:
Hi all,
there had been several attempts to implement a latency-based I/O
scheduler for native nvme multipath, all of which had its issues.
So time to start afresh, this time using the QoS framework
already present in the block layer.
It consists of two parts:
- a new 'blk-nlatency' QoS module, which is just a simple per-node
latency tracker
- a 'latency' nvme I/O policy
Using the 'tiobench' fio script with 512 byte blocksize I'm getting
the following latencies (in usecs) as a baseline:
- seq write: avg 186 stddev 331
- rand write: avg 4598 stddev 7903
- seq read: avg 149 stddev 65
- rand read: avg 150 stddev 68
Enabling the 'latency' iopolicy:
- seq write: avg 178 stddev 113
- rand write: avg 3427 stddev 6703
- seq read: avg 140 stddev 59
- rand read: avg 141 stddev 58
Setting the 'decay' parameter to 10:
- seq write: avg 182 stddev 65
- rand write: avg 2619 stddev 5894
- seq read: avg 142 stddev 57
- rand read: avg 140 stddev 57
That's on a 32G FC testbed running against a brd target,
fio running with 48 threads. So promises are met: latency
goes down, and we're even able to control the standard
deviation via the 'decay' parameter.
As usual, comments and reviews are welcome.
Changes to the original version:
- split the rqos debugfs entries
- Modify commit message to indicate latency
- rename to blk-nlatency
Hannes Reinecke (2):
block: track per-node I/O latency
nvme: add 'latency' iopolicy
John Meneghini (1):
nvme: multipath: pr_notice when iopolicy changes
MAINTAINERS | 1 +
block/Kconfig | 9 +
block/Makefile | 1 +
block/blk-mq-debugfs.c | 2 +
block/blk-nlatency.c | 389 ++++++++++++++++++++++++++++++++++
block/blk-rq-qos.h | 6 +
drivers/nvme/host/multipath.c | 73 ++++++-
drivers/nvme/host/nvme.h | 1 +
include/linux/blk-mq.h | 11 +
9 files changed, 484 insertions(+), 9 deletions(-)
create mode 100644 block/blk-nlatency.c
--
2.39.3
On Thu, May 09, 2024 at 04:43:21PM -0400, John Meneghini wrote:
> I'm re-issuing Hannes's latency patches in preparation for LSFMM
Hello John,
Just a small note.
Please don't reply-to the previous version of the series (v2), when sending
out a v3.
It creates "an unmanageable forest of references in email clients".
See:
https://www.kernel.org/doc/html/latest/process/submitting-patches.html#explicit-in-reply-to-headers
Instead just add the url to the v2 on lore.kernel.org.
See you at LSFMM!
Kind regards,
Niklas
>
> Changes since V2:
>
> I've done quite a bit of work cleaning up these patches. There were a
> number of checkpatch.pl problems as well as some compile time errors
> when config BLK_NODE_LATENCY was turned off. After the clean up I
> rebased these patches onto Ewan's "nvme: queue-depth multipath iopolicy"
> patches. This allowed me to test both iopolicy changes together.
>
> All of my test results, together with the scripts I used to generate these
> graphs, are available at:
>
> https://github.com/johnmeneghini/iopolicy
>
> Please use the scripts in this repository to do your own testing.
>
> Changes since V1:
>
> Hi all,
>
> there had been several attempts to implement a latency-based I/O
> scheduler for native nvme multipath, all of which had its issues.
>
> So time to start afresh, this time using the QoS framework
> already present in the block layer.
> It consists of two parts:
> - a new 'blk-nlatency' QoS module, which is just a simple per-node
> latency tracker
> - a 'latency' nvme I/O policy
>
> Using the 'tiobench' fio script with 512 byte blocksize I'm getting
> the following latencies (in usecs) as a baseline:
> - seq write: avg 186 stddev 331
> - rand write: avg 4598 stddev 7903
> - seq read: avg 149 stddev 65
> - rand read: avg 150 stddev 68
>
> Enabling the 'latency' iopolicy:
> - seq write: avg 178 stddev 113
> - rand write: avg 3427 stddev 6703
> - seq read: avg 140 stddev 59
> - rand read: avg 141 stddev 58
>
> Setting the 'decay' parameter to 10:
> - seq write: avg 182 stddev 65
> - rand write: avg 2619 stddev 5894
> - seq read: avg 142 stddev 57
> - rand read: avg 140 stddev 57
>
> That's on a 32G FC testbed running against a brd target,
> fio running with 48 threads. So promises are met: latency
> goes down, and we're even able to control the standard
> deviation via the 'decay' parameter.
>
> As usual, comments and reviews are welcome.
>
> Changes to the original version:
> - split the rqos debugfs entries
> - Modify commit message to indicate latency
> - rename to blk-nlatency
>
> Hannes Reinecke (2):
> block: track per-node I/O latency
> nvme: add 'latency' iopolicy
>
> John Meneghini (1):
> nvme: multipath: pr_notice when iopolicy changes
>
> MAINTAINERS | 1 +
> block/Kconfig | 9 +
> block/Makefile | 1 +
> block/blk-mq-debugfs.c | 2 +
> block/blk-nlatency.c | 389 ++++++++++++++++++++++++++++++++++
> block/blk-rq-qos.h | 6 +
> drivers/nvme/host/multipath.c | 73 ++++++-
> drivers/nvme/host/nvme.h | 1 +
> include/linux/blk-mq.h | 11 +
> 9 files changed, 484 insertions(+), 9 deletions(-)
> create mode 100644 block/blk-nlatency.c
>
> --
> 2.39.3
>
>