Hi,
A somewhat random perusal of PBS queue definitions on LCG CE nodes (via
globus-job-run and qstat) is showing me that most sites are configured
to have three queues:
short: 15m/2h
long: 12h/24h
infinite: 48h/72h
where the first number is the max CPUtime, and the second number is the
max wall clock time.
I am wondering if there is some sort of "default" partition of the
percentage of jobs allowed in each queue. e.g. something like:
* 20% or 5 nodes, whichever is less, in infinite
* 40% or 20 nodes, whichever is less, in short
* the rest in long.
If so, is it correct to assume that we (LHCb) should try to keep our
jobs to under 12h CPUtime and 24 hours wall clock time so they can fit
in the default "long" queues, rather than be relegated to the (possibly)
lower priority/smaller resource "infinite" queues?
Cheers,
Ian.
--
Ian Stokes-Rees [log in to unmask]
Particle Physics, Oxford http://www-pnp.physics.ox.ac.uk/~stokes
|