[torqueusers] maui + torque job start rate
josh at clusterresources.com
Wed Apr 1 09:34:59 MDT 2009
----- "Ling C. Ho" <ling at fnal.gov> wrote:
> Our single-np jobs take 3-6 seconds to start, when there are already
> jobs running on the worker
> nodes (configured as 2-8 nps). Even using qrun -a, it still take close
> to 2 seconds to start.
> Starting jobs on batch of "free" worker nodes is really fast, but most
> of the time we have some jobs
> already running on the worker nodes. Would Moab Asyncstart help in
> this case? Do the jobs actually
> get started, or are they just being pushed to Torque in a higher
They are actually being started. Like I said, 50 jobs/sec is the peak performance, but I do regularly see 20 jobs/sec. in our labs. Moab Asyncstart probably won't magically make jobs start faster on your system if qrun -a isn't helping a whole lot. There may be other parts of your config that are slowing things down.
More information about the torqueusers