On 12/28/05, Aritz Bastida <[EMAIL PROTECTED]> wrote:
> Hi everybody:
>
> I need to use pktgen for sending packets at very high speed to another
> machine, in order to test it under heavy network traffic. All my
> previous injection test were done with a dual Pentium III 800 MHz. As
> I needed a more powerful machine I got a Pentium 4 but the results are
> quite similar.
>
> The new machine specifications are:
>
> Processor:  Pentium 4  2.8 GHz
> Kernel:       2.6.13 / 2.6.14.5  (I've tested both of them)
> 2 NICs:      e1000 (Intel 82541GI/PI and Intel 82547GI)
> Bus:          PCI-X, I think (that's what it says in lspci, but I'm
> not sure how to find the bus
>                                       speed and all that)
>
> The results with pktgen are:
> pkt_size=100
> count=10000000
> clone_skb=0
> delay=0
>                           pktgen.packet_rate (pps): 170057
>                           pktgen.throughput (Mbps): 136
>                           pktgen.total_time (us): 58803762
>                           pktgen.work_time (us): 57141059
>                           pktgen.idle_time (us): 1662703
> --------------------------------------------------------------------------------------
> pkt_size=100
> count=10000000
> clone_skb=1000000
> delay=0
>                           pktgen.packet_rate (pps): 323602
>                           pktgen.throughput (Mbps): 258
>                           pktgen.total_time (us): 30902126
>                           pktgen.work_time (us): 29325060
>                           pktgen.idle_time (us): 1577066
>
> These results are similar to the ones I got with the Pentium III. I
> can't reach even 400kpps. With another machine, a dual AMD Opteron, I
> can send as fast as 650kpps but unfortunately that is the machine
> being tested. In the pktgen paper says "A single flow of 1.48Mpps is
> seen with a Xeon 2.67 GHz using a patched e1000 driver (64 byte
> packets)" [1]. Well, I don't know how much faster the Intel Xeon is,
> but I do have a fast machine, and I do have an e1000 NIC.
>
> So, if anyone can help me, please help. It's maybe that I need the
> patch for the e1000 to achieve those high numbers. However, I can't
> find that patch, so if anyone knows about it, please tell me where to
> find it.
>
> Thank you. Regards
> Aritz
>
> [1] Paper from Linux-Kongress in Erlangen 2004.
>      
> ftp://robur.slu.se/pub/Linux/net-development/pktgen-testing/pktgen_paper.pdf

The patches mentioned are at this site above, in the net-development
directory, if i'm not mistaken.

Also, you'll get a big improvement from running a UP kernel with packetgen.

reply with output of lspci -vvv -xxx -s <your ethernet adapter's slot>
and ethtool -d ethX output and we can tell your bus speed from those.

also send ethtool -S ethX output after running your pktgen, we're
looking to see if you need to turn off flow control.

running an x86_64 kernel will help your pktgen numbers on the P4 as
well.  Don't know why fully yet.
-
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Reply via email to