[ovs-dev] [PATCH 00/11] RFC: Improvements in upcall processing code path.

[Ben Pfaff]

"netperf -t TCP_CRR" is a reasonable simulation of apachebench but
much easier to set up.

On Tue, Feb 12, 2013 at 03:47:50PM -0800, Ethan Jackson wrote:
> FWIW: It might be useful to check if this series has an impact on
> something like apachebench which I would expect benefit more from
> batching, but be a little bit less specialized than either
> ovs-benchmark or udpping.
> 
> Ethan
> 
> 
> On Tue, Feb 12, 2013 at 2:03 PM, Ben Pfaff <blp at nicira.com> wrote:
> > On Mon, Feb 11, 2013 at 04:46:16PM +0200, Jarno Rajahalme wrote:
> >> - 'udpping' sends 10000000 UDP packets with a unique source and destination
> >>   port pair to br0 in bursts of 97 packets.  Between bursts it nanosleeps
> >>   for 50 nanoseconds.  This results into ~80% CPU (core) utilization.
> >
> > Why are the bursts 97 packets long?
> >
> > It would be great if you could supply the udpping program.  Ideally,
> > you'd add it as a mode to the ovs-benchmark program in the utilities
> > directory, but a quick-and-dirty C program would work too.
> > _______________________________________________
> > dev mailing list
> > dev at openvswitch.org
> > http://openvswitch.org/mailman/listinfo/dev