Re: [TEST/REVIEW] CPU accounting patches

On Friday 27 January 2006 21:16, Brooks Davis wrote:

[snip]

> I agree as well.  Certainly if we were charging for use of our cluster,
> this is what we'd want.  While I probably wouldn't run powerd on the
> cluster, I and thinking about seeing if I can step down the CPU speed
> when there aren't any queued jobs on the machine.  That could save
> significant power some of the time (I'm in the process of upgrading the
> cluster portion of our server room to install 300KVA (~KW) of power and
> plan to use it all within a year or two).
>
> Once we have the infrastructure to deal with this correctly, an
> intresting test for someone to run would be to look at disk and memory
> bound applications at different CPU speeds.  I suspect you'd find that
> while wallclock increased at lower CPU speeds, cpu cycles would decrease
> for many workloads because the relative bandwidth of storage and maybe
> memory would increase.
>
> -- Brooks

Just to give the discussion a different angle - Across most very large 
midrange estates - the expectated maximum use of systems when measured over a 
full year averages is app. 8% - The % being sligthly tricky because it 
includes name servers, NIS etc. If one refines the data you find that the 
best case looks like a maximum use of around 16% - 20% - The measurements 
made over the entire estate of a couple of ITO's and hardware vendors - e.g. 
based upon 50,000+ servers - I that view the ability to account effectively 
for Use/Speed etc. are somewhat more interessting