All of lore.kernel.org
 help / color / mirror / Atom feed
From: Greg Banks <gnb@melbourne.sgi.com>
To: "Lever, Charles" <Charles.Lever@netapp.com>
Cc: James Pearson <james-p@moving-picture.com>, nfs@lists.sourceforge.net
Subject: Re: Different NFSSVC_MAXBLKSIZE for udp and tcp clients?
Date: Thu, 27 Nov 2003 11:08:06 +1100	[thread overview]
Message-ID: <3FC54066.C04EBA49@melbourne.sgi.com> (raw)
In-Reply-To: 482A3FA0050D21419C269D13989C6113020AC79F@lavender-fe.eng.netapp.com

"Lever, Charles" wrote:
> 
> here's a vote for leaving these values variable rather than
> fixed on the server side.  48KB maximum for UDP is great on
> a clean network, but some folks would probably prefer setting
> it to 8KB to ensure clients on more realistically congested
> networks retain reasonable UDP performance.

Agreed.

> is there any evidence to show that a 4MB TCP maximum will
> have benefits over something smaller, like say 1MB ?  

None at all on Linux.  I mention that specific number only
because it's the hardcoded maximum on IRIX.

However there is evidence that 8K isn't enough.  In particular,
it's smaller than an Altix client's PAGE_CACHE_SIZE so all writes
between two Altix boxes go synchronous.

> [...] there's probably a sweet spot where the sum
> of both per-op and per-byte costs reach a minimum.  

Yes.

> as
> far as i can tell that value is around 16KB in the current
> Linux client implementations.

The value will be higher for Altix clients and servers, which
have more RAM, CPU, and larger pages than the machines you're
probably thinking of.

Greg.
-- 
Greg Banks, R&D Software Engineer, SGI Australian Software Group.
I don't speak for SGI.


-------------------------------------------------------
This SF.net email is sponsored by: SF.net Giveback Program.
Does SourceForge.net help you be more productive?  Does it
help you create better code?  SHARE THE LOVE, and help us help
YOU!  Click Here: http://sourceforge.net/donate/
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

  reply	other threads:[~2003-11-27  0:08 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2003-11-26 14:20 Different NFSSVC_MAXBLKSIZE for udp and tcp clients? Lever, Charles
2003-11-27  0:08 ` Greg Banks [this message]
2003-11-27  0:53   ` Trond Myklebust
  -- strict thread matches above, loose matches on Subject: below --
2003-11-25 17:34 James Pearson
2003-11-26  1:08 ` Greg Banks
2003-11-26  1:40   ` Trond Myklebust
2003-11-26  1:51     ` Greg Banks
2003-11-26  2:00       ` Trond Myklebust
2003-11-26 10:43   ` James Pearson
2003-11-26 22:15     ` Neil Brown
2003-11-27  0:21       ` Greg Banks
2003-11-27  0:35         ` Neil Brown
2003-11-27  1:33           ` Greg Banks

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3FC54066.C04EBA49@melbourne.sgi.com \
    --to=gnb@melbourne.sgi.com \
    --cc=Charles.Lever@netapp.com \
    --cc=james-p@moving-picture.com \
    --cc=nfs@lists.sourceforge.net \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.