[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] limiting memory use of condor jobs



This feature looks quite useful.  The question is.. is this
something that can be automatically appended to all job classads
or  do you have to trust the user to put it in their submit file,
if so, how?

Steve Timm


On Fri, 10 Feb 2006, Alexander Lincoln Read wrote:

+JobMemoryLimit = 500000
+JobDiskLimit = 10000
+JobTimeLimit = 3600
+JobCpuLimit = 1800
periodic_remove = DiskUsage > JobDiskLimit ||  ImageSize >
JobMemoryLimit || RemoteWallClockTime > JobTimeLimit || RemoteUserCpu >
JobCpuLimit

will remove jobs that either
1) use more than 10 MB of disk
2) use more than 500 MB of memory
3) run for more than an hour
4) use more than 30 minutes of CPU

I made (3) and (4) different just for fun, but for an application with a
lot of disk reading or
networking it could  make sense.

Cheers, Alex





Hans Wenzel wrote:

Dear all


Is there a way  to periodic_remove a Job once the memory use is above
some threshold?
like it is done in the diskusage example below?

thanks in advance

Hans


In the job requirements
+JobDiskLimit = 10000
periodic_remove = DiskUsage > JobDiskLimit
should prevent the job from using more than e.g. 10 MB in
the session directory. I wouldn't know how to get Condor to
enforce this for a particular file - you would probably have to
build a command wrapper and check the file size in a separate
process if this is important.

Cheers, Alex


_______________________________________________
Condor-users mailing list
Condor-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/condor-users



_______________________________________________
Condor-users mailing list
Condor-users@xxxxxxxxxxx
https://lists.cs.wisc.edu/mailman/listinfo/condor-users


--
------------------------------------------------------------------
Steven C. Timm, Ph.D  (630) 840-8525  timm@xxxxxxxx  http://home.fnal.gov/~timm/
Fermilab Computing Div/Core Support Services Dept./Scientific Computing Section
Assistant Group Leader, Farms and Clustered Systems Group
Lead of Computing Farms Team