[GE users] parallel job woes

ah_sunsource ahaupt at ifh.de
Tue Jun 30 09:52:27 BST 2009


Hi *,

we observe a problem with huge parallel jobs (>50 subtasks). Each
parallel task requires round about 500MB h_vmem (h_vmem is configured as
consumable host resource). So that's the actual memory requirement. But
starting the subtasks via qrsh in the master process uses >50MB virtual
memory for every connection. That's why the job gets killed.

Is there a way to specify a different memory requirement for the master
task? Or maybe switch off memory enforcing for the master task as this
is actually SGE internal mem consumption?

Thanks & Cheers,
Andreas

-- 
| Andreas Haupt             | E-Mail: andreas.haupt at desy.de
|  DESY Zeuthen             | WWW:    http://www-zeuthen.desy.de/~ahaupt
|  Platanenallee 6          | Phone:  +49/33762/7-7359
|  D-15738 Zeuthen          | Fax:    +49/33762/7-7216

------------------------------------------------------
http://gridengine.sunsource.net/ds/viewMessage.do?dsForumId=38&dsMessageId=204554

To unsubscribe from this discussion, e-mail: [users-unsubscribe at gridengine.sunsource.net].



More information about the gridengine-users mailing list