[torqueusers] Delayed Job Execution
knielson at adaptivecomputing.com
Tue Oct 5 09:00:28 MDT 2010
On 10/05/2010 08:59 AM, Mark A. White wrote:
> Has anyone experienced a problem with PBS/Torque delaying execution of
> jobs, even when adequate resources are available? In my experience,
> this is not a repeatable problem, sometimes 100 jobs will be launched
> simultaneously, and another time 5 of fifty will be delayed, with no
> apparent difference in circumstances.
> Any ideas how to debug this problem?
> On Wed, 2010-09-29 at 17:24 -0500, Mark A. White wrote:
>> A user is experience an apparently random issue with the Torque/PDB
>> scheduler delaying execution of jobs, even when adequate resources
>> are available.
>> He will submit a batch of 50 or 100 small jobs simultaneously (in a
>> single script). Sometimes all 50 or 100 jobs will run. At other
>> times several jobs will be held until the other finish (just a few
>> minutes). This does not seem to relate to the available resources:
>> there is not a conflict with other jobs queued or running.
>> Has anyone else experienced this, know what the cause and solution to
>> the problem might be?
>> torque 2.1.9-1
>> CentOS 5.2
>> 120 processor cluster (15 boxes, 64 bit dual quad-core opterons)
>> Yours sincerely,
>> Mark A. White, Ph.D.
>> Associate Professor of Biochemistry and Molecular Biology,
>> Manager, Sealy Center for Structural Biology and Molecular Biophysics
>> X-ray Crystallography Laboratory,
>> Basic Science Building, Room 6.660 C
>> University of Texas Medical Branch
>> Galveston, TX 77555-0647
>> Tel. (409) 747-4747
>> Cell. (281) 734-3614
>> Fax. (409) 747-1404
> Yours sincerely,
> Mark A. White, Ph.D.
> Associate Professor of Biochemistry and Molecular Biology,
> Manager, Sealy Center for Structural Biology and Molecular Biophysics
> X-ray Crystallography Laboratory,
> Basic Science Building, Room 6.660 C
> University of Texas Medical Branch
> Galveston, TX 77555-0647
> Tel. (409) 747-4747
> Cell. (281) 734-3614
> Fax. (409) 747-1404
> torqueusers mailing list
> torqueusers at supercluster.org
What are you using as a scheduler?
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the torqueusers