Re: [NFS] [PATCH] knfsd: Correct reserved reply space for read requests.

From: Neil Brown
Date: Sun Apr 02 2006 - 23:47:58 EST


On Sunday April 2, eshel@xxxxxxxxxxxxxxx wrote:
> nfs-admin@xxxxxxxxxxxxxxxxxxxxx wrote on 04/02/2006 06:17:17 PM:
>
> > On Thursday March 30, eshel@xxxxxxxxxxxxxxx wrote:
> > > Hi Neil
> > > Can we use this opportunity to change NFSSVC_MAXBLKSIZE from 32K to
> 64K to
> > > match RPCSVC_MAXPAYLOAD. It makes real difference in I/O performance
> (we
> > > will still be saving half the space we used to allocate :).
> > > Thanks, Marc.
> >
> > Maybe... but why not 128K ??
>
> Yes, It would be nice to be able to match the Linux client side that can
> go to 1MB.
>
> > There is certainly room to increase NFSSVC_MAXBLKSIZE, but I feel that
> > it needs to be done together with a more detailed look at consequences
> > in the network layer, particularly TCP window sizes. I wouldn't mind
> > making a CONFIG_ tunable without that detailed look, but making it a
> > fixed change I feel less comfortable about.
>
> Like you said it will need match more work to do it right and also not
> waste space for all RPC request to which the maximum possible size is
> allocated up front. But until than way not take advantage of this minor
> change that can give us significant performance improvement. I run with
> NFSSVC_MAXBLKSIZE set to 64K (the only change I made) and saw 10%-15%
> improvement for NFS reads. Is there anyone out there that is looking at
> making this improvement ?

It was a fairly soft reservation of space, and we never actually tried
to use it all.
I don't see a clear problem with just making the change you mention,
but until I have thought through the implications (or until someone
else has and convinces me) I don't feel comfortable making the change.

It's on my todo list, but that doesn't promise a lot :-(
If someone else wants to try to convince me it is safe, and mention
tcp window sizes a couple of time, I'll probably be happy with it.

NeilBrown
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/