[P4-dev] iperf on heavy hitter example

mostafa ammar mostafaammar79 at gmail.com
Sun Oct 15 14:28:42 EDT 2017


Dear All,

I tried rebuilding the bmv2 switch with  ./configure --disable-logging-macros
--disable-elogger

throughput reached 80Mbps and during running syn attack from host to
another host 35 Mbps.

It became better than the old bmv2 .

Thanks.


On Wed, Oct 11, 2017 at 9:23 PM, Antonin Bas <antonin at barefootnetworks.com>
wrote:

> This has been covered before: http://lists.p4.org/
> pipermail/p4-dev_lists.p4.org/2017-February/000715.html
> While production performance has never been a design goal of bmv2, there
> is no reason why you shouldn't at least get a couple hundred Mbps with such
> a simple P4 program if you use the appropriate compile-time & runtime flags.
>
> On Wed, Oct 11, 2017 at 11:07 AM, Mostafa Abdallah. Ammar <
> mostafaammar at aast.edu> wrote:
>
>>     ​Dear All,
>>
>>
>> I am trying to measure  throughput between 2 hosts connected on P4
>> simple_Switch, I used the iperf tool and used the heavy_hitter example. the
>> measured throughput was 36.7 Mbps
>>
>> however when trying the same experiment on openvswitches the throughput
>> was a lot higher.
>>
>>
>>
>>
>> Best Regards,
>>
>> Eng. Mostafa Abdallah Ammar,Msc.
>> Head of Information Security Department
>> CCIE security #23971
>> Arab Academy For Science And Technology & maritime Transport
>> Computer Networks & Data Center (CNDC)
>> Mobile: 002 01001983674
>>
>> _______________________________________________
>> P4-dev mailing list
>> P4-dev at lists.p4.org
>> http://lists.p4.org/mailman/listinfo/p4-dev_lists.p4.org
>>
>
>
>
> --
> Antonin
>
> _______________________________________________
> P4-dev mailing list
> P4-dev at lists.p4.org
> http://lists.p4.org/mailman/listinfo/p4-dev_lists.p4.org
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.p4.org/pipermail/p4-dev_lists.p4.org/attachments/20171015/def8901e/attachment-0002.html>


More information about the P4-dev mailing list