LHC Computer Grid - Rollout
> [mailto:[log in to unmask]] On Behalf Of Alessandro Paolini
said:
> if each of your WN has 8 core, 226*8=1808 is the right
> number; but, if
> I've well understood your case, if on some WN the overbooking
> makes to
> see to the O.S. more "computing units", you should take them into
> account in the "LogicalCPU number"
No, as I just said the LogicalCPUs are representing your physical
hardware so they don't depend on the batch system configuration. The job
slot numbers are reported separately by the dynamic info providers in
the CE and VOView objects.
> actually you should publish "Cores=4,Benchmark=54.90-HEP-SPEC06",
> because 4 is the number of cores per cpu on the typical WN
I missed that in the previous mail, but indeed the Cores number is cores
per (physical) CPU and not per WN. At the moment I think that's mainly
informational, i.e. just so we can see what kind of machines the sites
are deploying. For the hardware accounting the important things are
LogicalCPUs (total number of cores), HEPSPEC (benchmark per core) and
the VO shares (fraction of LogicalCPUs * HEPSPEC allocated to each VO).
Stephen
--
Scanned by iCritical.
|