netdev
[Top] [All Lists]

Re: 2.6.7 tulip performance (with NAPI)

To: Lennert Buytenhek <buytenh@xxxxxxxxxxxxxx>
Subject: Re: 2.6.7 tulip performance (with NAPI)
From: P@xxxxxxxxxxxxxx
Date: Fri, 08 Oct 2004 10:25:13 +0100
Cc: Robert Olsson <Robert.Olsson@xxxxxxxxxxx>, Ben Greear <greearb@xxxxxxxxxxxxxxx>, "'netdev@xxxxxxxxxxx'" <netdev@xxxxxxxxxxx>
In-reply-to: <20041007234413.GC24969@xi.wantstofly.org>
References: <41633174.7070805@candelatech.com> <16740.17875.574967.11417@robur.slu.se> <41646587.7070401@candelatech.com> <41649425.1010102@candelatech.com> <16741.45331.743721.822091@robur.slu.se> <20041007234413.GC24969@xi.wantstofly.org>
Sender: netdev-bounce@xxxxxxxxxxx
User-agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040124
Lennert Buytenhek wrote:
On Thu, Oct 07, 2004 at 11:11:47PM +0200, Robert Olsson wrote:


I'm interested if anyone has done any pktgen performance tests with
w. S2IO or other 10G card we need to upgrade the lab equipment. Both 64 byte pkts and MTU sized pkts is interesting. Anyone?


If you're interested in raw pps numbers and willing to use something
other than pktgen to attain them, you might want to look into the intel
IXP-based boards.  You actually have to write some microcode and upload
it to an on-chip processor array, but you can easily sustain wirespeed
at 4Gbps (IXP2400-based) or 10Gbps (IXP2800-based.)

Good suggestion. This is exactly what I'm doing on a (2400 based) radisys enp-2611 board. Faced with the cost of a smartbits it's a no brainer for me. I haven't written the µ code yet btw.

Pádraig.

<Prev in Thread] Current Thread [Next in Thread>