Subject: | |
From: | |
Reply To: | |
Date: | Sat, 11 Jul 2009 07:51:41 +0100 |
Content-Type: | TEXT/PLAIN |
Parts/Attachments: |
|
|
On Fri, 10 Jul 2009, Eli Dart wrote:
> It appears that Scientific Linux is used by many science communities,
> but in particular by the HEP community. The science community often has
> significant bulk data movement requirements that are outside the
> capabilities of the default network tuning parameters of most Linux
> distributions.
>
> Would the Scientific Linux community consider changing the network
> tuning defaults for future releases?
I support Troy's decision to leave the default,
but to have a package which tunes things to some agreed alternative.
> Note that with today's TCP autotuning and modern congestion recovery
> algorithms, one need not set up particular TCP parameters on a
> per-destination basis. One need only give TCP autotuning enough buffer
> space to do its work and ensure that a modern congestion recovery
> algorithm is used (the default in Linux 2.6 has been cubic for a while,
> though 2.6.18 has bugs in cubic that significantly damage performance so
> for 2.6.18 one should use htcp instead).
Seems like fixing cubic in 2.6.18 would be desirable.
Would back-porting the fixes be viable ?
If it is too much of a change for SL, you could ask TUV (RedHat) ?
--
Dr. Andrew C. Aitchison Computer Officer, DPMMS, Cambridge
[log in to unmask] http://www.dpmms.cam.ac.uk/~werdna
|
|
|