Mailing List Archives
Authenticated access
|
|
|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Condor-users] wait times
- Date: Fri, 20 Apr 2012 17:12:38 -0400 (EDT)
- From: Tim St Clair <tstclair@xxxxxxxxxx>
- Subject: Re: [Condor-users] wait times
Short answer is to use Rank | Requirements.
http://research.cs.wisc.edu/condor/manual/v7.7/2_5Submitting_Job.html#SECTION00352000000000000000
possibly based on Target.LoadAverage
Cheers,
Tim
----- Original Message -----
> From: "Donald C Shrum" <DCShrum@xxxxxxxxxxxxx>
> To: condor-users@xxxxxxxxxxx
> Sent: Friday, April 20, 2012 3:51:33 PM
> Subject: [Condor-users] wait times
>
> Our condor cluster has about 720 slots. Our servers are dedicated to
> condor.
>
> We have several schedulers running. When I submit a test job from
> one submit node my condor process runs almost immediately.
>
> Submitted from another node the process runs for in excess of 30
> minutes.
>
> Nothing in the logs shows a problem. My inclination is that this is
> related to the processing node running the job and not the submit
> node.
>
> >From submit node slow I can see the job is sent to processing node
> >21 where it seems to just sit.
>
> >From submit node fast the job goes to processing node 130 where it
> >runs immediately.
>
> Given that each 'node' contains 12 slots (one for each cpu) does
> condor assign priority is some way? On submit node 21 would it make
> slot2 wait for slot1 to complete before running?
>
> Any pointers to debug this would be appreciated!
>
> --Don
> Florida State University HPC
>
> _______________________________________________
> Condor-users mailing list
> To unsubscribe, send a message to condor-users-request@xxxxxxxxxxx
> with a
> subject: Unsubscribe
> You can also unsubscribe by visiting
> https://lists.cs.wisc.edu/mailman/listinfo/condor-users
>
> The archives can be found at:
> https://lists.cs.wisc.edu/archive/condor-users/
>