This is perhaps a bit difficult to explain, but --topspeed will always
be significantly faster then the maximum value for --pps with the
current code.  Using anything but --topspeed introduces some
additional calculations to figure out sleep times + the overhead of
sleeping which depending on your platform may or may not be very
accurate.  Using --mbps and --multiplier are even worse then --pps
since even more calculations are necessary and I can't use weighted
averages.

I'm currently working on some code which specifically address this
issue so that higher packet rates are possible when using --pps, but
it never will equal --topspeed.

Also, I've no idea what version of glibc your box has, but older
versions have an issue which causes bad timings.  This issue was fixed
in the latest version of tcpreplay.

And of course if you haven't recently read the sections in the
Tcpreplay manual about timing methods and tuning for high performance,
you should refresh your memory:

http://tcpreplay.synfin.net/trac/wiki/tcpreplay#ChoosingaTimingMethod

You might find trying a different timing method helps alot.

-- 
Aaron Turner
http://synfin.net/
http://tcpreplay.synfin.net/ - Pcap editing and replay tools for Unix & Windows
They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety.  -- Benjamin Franklin


On Mon, Nov 10, 2008 at 11:37 PM, Long Lei <[EMAIL PROTECTED]> wrote:
> Hi,
>
>
>
> 1, My OS is Linux ubuntu 2.6.27-7-server #1 SMP Fri Oct 24 07:37:55 UTC 2008
> i686 GNU/Linux
>
> 2, Hardware: 4G RAM, AMD Athlon 64 Dual-core 1*2.6G
>
> 3, tcpreplay -V
>
> tcpreplay version: 3.3.1 (build 2033)
>
> Copyright 2001-2008 by Aaron Turner <aturner at synfin dot net>
>
> Cache file supported: 04
>
> Not compiled with libnet.
>
> Compiled against libpcap: 0.9.8
>
> 64 bit packet counters: enabled
>
> Verbose printing via tcpdump: enabled
>
> Packet editing: disabled
>
> Fragroute engine: disabled
>
>
>
> 3, When I ran "tcpreplay --topspeed --intf1=eth0 all.pcap", the result was:
>
> sending out eth0
>
> processing file: all.pcap
>
> Actual: 253586 packets (28122441 bytes) sent in 0.96 seconds
>
> Rated: 29134243.1 bps, 222.28 Mbps/sec, 262709.64 pps
>
>
>
> 4,When I ran "tcpreplay --pps=200000 --intf1=eth0 all.pcap", the result was:
>
> sending out eth0
>
> processing file: all.pcap
>
> Actual: 135215 packets (15026874 bytes) sent in 1.78 seconds
>
> Rated: 8426957.1 bps, 64.29 Mbps/sec, 75827.55 pps
>
>
>
> It seemed that "--pps" didn't work. Would you like to help me out? Thank you
> very much!

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Tcpreplay-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/tcpreplay-users
Support Information: http://tcpreplay.synfin.net/trac/wiki/Support

Reply via email to