svn commit: r240742 - head/sys/net

Nikolay Denev ndenev at gmail.com
Fri Oct 5 15:46:55 UTC 2012


On Oct 5, 2012, at 6:11 PM, Gleb Smirnoff <glebius at FreeBSD.org> wrote:

>  Nikolay,
> 
> On Fri, Oct 05, 2012 at 05:11:12PM +0300, Nikolay Denev wrote:
> N> With both modules I was able to saturate the four GigE interfaces, and got 
> N> about ~3.72 Gbits/sec total according to iperf, systat -ifstat showed
> N> about 116MB/s per each interface.
> N> 
> N> However I'm seeing slightly different CPU stat graphs [1], the difference is not big,
> N> but with the new if_lagg(4) driver, when the machine is acting as client I'm
> N> seeing slightly higher system CPU time, and about the same interrupt, while
> N> when acting as server both system and interrupt are slightly lower.
> N> But please note that these tests were not very scientifically correct.
> N> When the server is available again I might be able to perform several runs and
> N> do a proper comparison.
> 
> Do I understand correct, that in the above testing "server" means transmitting
> traffic and "client" is receiving traffic?
> 
> -- 
> Totus tuus, Glebius.

Actually with iperf the server is more like a sink, and the client sends data to the server.
Here's what's in the man page :

To perform an iperf test  the  user
       must establish both a server (to discard traffic) and a client (to gen-
       erate traffic).



More information about the svn-src-all mailing list