[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Pine.LNX.4.64.0903302224120.13385@ask.diku.dk>
Date: Mon, 30 Mar 2009 22:34:14 +0200 (CEST)
From: Jesper Dangaard Brouer <hawk@...u.dk>
To: Eric Dumazet <dada1@...mosbay.com>
Cc: netdev <netdev@...r.kernel.org>,
Netfilter Developers <netfilter-devel@...r.kernel.org>
Subject: Re: [PATCH] netfilter: finer grained nf_conn locking
On Mon, 30 Mar 2009, Eric Dumazet wrote:
> Jesper Dangaard Brouer a écrit :
>>
>>> Eric Dumazet wrote:
>>>> "tbench 8" results on my 8 core machine (32bit kernel, with
>>>> conntracking on) : 2319 MB/s instead of 2284 MB/s
>>
>> How do you achieve this impressing numbers?
>> Is it against localhost? (10Gbit/s is max 1250 MB/s)
>>
>
> tbench is a tcp test on localhost yes :)
I see!
Using a Sun 10GbE NIC I was only getting a throughput of 556.86 MB/sec
with 64 procs (between an AMD Phenom X4 and a Core i7). (Not tuned multi
queues yet ...)
Against localhost I'm getting (not with applied patch):
1336.42 MB/sec on my AMD phenom X4 9950 Quad-Core Processor
1552.81 MB/sec on my Core i7 920 (4 physical cores, plus 4 threads)
2274.53 MB/sec on my dual CPU Xeon E5420 (8 cores)
> Good to test tcp stack without going to NIC hardware
Yes true, but this also stresses the process scheduler, I'm seeing around
800.000 context switches per sec on the Dual CPU Xeon system.
Cheers,
Jesper Brouer
--
-------------------------------------------------------------------
MSc. Master of Computer Science
Dept. of Computer Science, University of Copenhagen
Author of http://www.adsl-optimizer.dk
-------------------------------------------------------------------
Powered by blists - more mailing lists