Hi,
I compiled NFS utils 1.0.6 for my Asus WL-HDD NAS device and I compiled the
2.4.27 kernel with support for up to 64Kb values (const.h: NFSSVC_MAXBLKSIZE
65536). It all works pretty good, but unfortunately if I try mounting with
rsize/wsize >32Kb the mount is always rsize=wsize=32Kb. I'm quite keen to
mount with larger wsize/rsize values as I expect better performance. How can
I accomplish larger rsize/wsize values?
Cheers,
JockyW
PS: for mounting I use mount from busybox 1.0.0
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://productguide.itmanagersjournal.com/
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs
Is it safe to assume you've enabled jumbo frames?
Without jumbo frames, a huge packet size isn't necessarily going to make
things faster...
On Thu, 2004-12-09 at 23:48 +0000, Jocky Wilson wrote:
> Hi,
>
> I compiled NFS utils 1.0.6 for my Asus WL-HDD NAS device and I compiled the
> 2.4.27 kernel with support for up to 64Kb values (const.h: NFSSVC_MAXBLKSIZE
> 65536). It all works pretty good, but unfortunately if I try mounting with
> rsize/wsize >32Kb the mount is always rsize=wsize=32Kb. I'm quite keen to
> mount with larger wsize/rsize values as I expect better performance. How can
> I accomplish larger rsize/wsize values?
>
> Cheers,
> JockyW
>
> PS: for mounting I use mount from busybox 1.0.0
>
>
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://productguide.itmanagersjournal.com/
> _______________________________________________
> NFS maillist - [email protected]
> https://lists.sourceforge.net/lists/listinfo/nfs
>
No, my LAN is fast ethernet and the NAS and PCs are in 100 full duplex. My
streaming client even has a 10mb half duplex nic. I noticed a consistent
write performance increase when changing wsize from 8192->16384->32768
I did increase rmem and wmem values from 32767 (yes) to 65536, which doesn't
seem to make a difference.
JockyW
>From: Dan Stromberg <[email protected]>
>Date: Thu, 09 Dec 2004 15:52:23 -0800
>
>
>Is it safe to assume you've enabled jumbo frames?
>
>Without jumbo frames, a huge packet size isn't necessarily going to make
>things faster...
>
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://productguide.itmanagersjournal.com/
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs
In http://nfs.sourceforge.net/nfs-howto/performance.html#PACKET-AND-NETWORK
it is written:
"Note that NFS Version 2 is limited to a maximum of 8K, regardless of the
maximum block size defined by NFSSVC_MAXBLKSIZE; Version 3 will support up
to 64K, if permitted"
My question is what requirements must be fullfilled such that NFSv3 daemon
will permit 64k blocksizes?
Cheers,
JockyW
-------------------------------------------------------
SF email is sponsored by - The IT Product Guide
Read honest & candid reviews on hundreds of IT Products from real users.
Discover which products truly live up to the hype. Start reading now.
http://productguide.itmanagersjournal.com/
_______________________________________________
NFS maillist - [email protected]
https://lists.sourceforge.net/lists/listinfo/nfs