ndt-users - Re: gigabit slower than fast-e
Subject: ndt-users list created
List archive
- From: Bill Abbott <>
- To: Jim Cotton <>
- Cc: Matthew J Zekauskas <>,
- Subject: Re: gigabit slower than fast-e
- Date: Fri, 07 Mar 2008 15:07:14 -0500
I don't think it's this; the same two servers could transfer at much higher speeds several months ago, when I did the original tuning.
Jim Cotton wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
I have seen several Dell servers on campus that exhibited such a problem
with their on-board chipset.
There is a problem with the Broadcom chip set below a certain
revision... If you get the right driver it will work around it.
Jim Cotton
Bill Abbott wrote:
I've tried tcp send and receive (max) buffers from 64k up to 8 MB, no
improvement.
Smaller buffers are more consistent, larger buffers tend to go up then
hit zero, then climb again.
The current settings are:
$ cat /proc/sys/net/ipv4/tcp_rmem
4096 87380 8174760
$ cat /proc/sys/net/ipv4/tcp_wmem
4096 16384 8131072
I've also modified a bunch of other stuff, but it didn't make a
difference so I set it back. ECN, /proc/sys/net/core/rmem_max and
wmem_max, netdev_max_backlog, etc.
Matthew J Zekauskas wrote:
To follow up to myself...
* The Internet2 servers are all gigE, and all 9000-byte MTU capable.
* Another thing you could do is tune for the bottleneck instead of
gigabit..
150 Mbps = 150*1000*1000 bits * 0.07 ms RTT = 10500000 bits
/8 = 1312500 bytes /(1024*1024) = 1.26 megabytes.
or
100 Mbps is 850K
and see if that improves things. It may be that something can't
handle even that much back-to-back when they are sent at 1000Mbps.
There are also bwctl (<http://e2epi.internet2.edu/bwctl/>) servers at
Internet2 router nodes, for a different way to split the problem
(essentially run iperf to the core).
--Matt
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.4-svn0 (GNU/Linux)
Comment: Using GnuPG with SUSE - http://enigmail.mozdev.org
iD8DBQFH0OaH2MvRXu0NtoARAtTrAJ0Z+dDKVvI9i3SYAwIBPeUhu/KqZACgqk2H
RltkpeLYhKoJYkh0BlWF7Ow=
=V7Ah
-----END PGP SIGNATURE-----
- gigabit slower than fast-e, abbott, 03/06/2008
- Re: gigabit slower than fast-e, Matthew J Zekauskas, 03/06/2008
- Re: gigabit slower than fast-e, Matthew J Zekauskas, 03/06/2008
- Re: gigabit slower than fast-e, Bill Abbott, 03/06/2008
- Re: gigabit slower than fast-e, Jim Cotton, 03/07/2008
- Re: gigabit slower than fast-e, Bill Abbott, 03/07/2008
- Re: gigabit slower than fast-e, Jim Cotton, 03/07/2008
- Re: gigabit slower than fast-e, Bill Abbott, 03/06/2008
- Re: gigabit slower than fast-e, Bill Abbott, 03/06/2008
- Re: gigabit slower than fast-e, Richard Carlson, 03/07/2008
- Re: gigabit slower than fast-e, Bill Abbott, 03/07/2008
- Re: gigabit slower than fast-e, Matt Mathis, 03/07/2008
- Re: gigabit slower than fast-e, Richard Carlson, 03/07/2008
- Re: gigabit slower than fast-e, Jeremy Schafer, 03/07/2008
- Re: gigabit slower than fast-e, Richard Carlson, 03/07/2008
- Re: gigabit slower than fast-e, Matt Mathis, 03/07/2008
- Re: gigabit slower than fast-e, Jeremy Schafer, 03/07/2008
- Re: gigabit slower than fast-e, Richard Carlson, 03/07/2008
- Re: gigabit slower than fast-e, Richard Carlson, 03/07/2008
- Re: gigabit slower than fast-e, Matthew J Zekauskas, 03/06/2008
- Re: gigabit slower than fast-e, Peter Van Epp, 03/06/2008
- Re: gigabit slower than fast-e, Bill Abbott, 03/06/2008
- Re: gigabit slower than fast-e, Matthew J Zekauskas, 03/06/2008
Archive powered by MHonArc 2.6.16.