[netperf-talk] Some question about CPU utilization

[Rick Jones]

Benjamin Thery wrote:
> Rick,
> 
> Your advices were really valuable.
> They helped me improve the situation a lot. Binding netperf to CPU,
> killing all unneeded services, increasing iterations,....
> 
> But still, I thought the reported CPU utilization was surprisingly low
> and one of the test was impossible to stabilize: UDP_RR (variations up
> to 500%, CPU utilization near 0%).
> 
> So, I thought "let's check the timer frequency configured in the kernel
> to see if a higher frequency helps netperf to calculate a more fine 
> grained CPU utilization", then I discovered my kernel was configured
> with CONFIG_NO_HZ=y : "Tickless System (Dynamic Ticks)".  8-|
> 
> I switched it to 'no', rebuilt, rebooted... and netperf worked like a
> charm then. No more variations reported for CPU utilization.
> I could even change my confidence range from 95,10 to 95,5.
> 
> Do you think CONFIG_NO_HZ=y can have an impact on how netperf uses
> /proc/stat to compute CPU utilization?

Netperf will still try to use it the same - but the usefulness of the 
information therein might be called into question.  If you feel up to 
it, mentioning it to your distro provider or upstream would be goodness.

happy benchmarking,

rick jones