Mailing List Archives
Authenticated access
|
|
|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Condor-users] Limiting number of concurrently running jobsfrom a cluster.
- Date: Tue, 26 Aug 2008 14:49:02 -0500 (CDT)
- From: "R. Kent Wenger" <wenger@xxxxxxxxxxx>
- Subject: Re: [Condor-users] Limiting number of concurrently running jobsfrom a cluster.
On Tue, 26 Aug 2008, Ian Chesal wrote:
I was going to answer with "use DAGMan" as well but then I got to
reading. Specifically:
http://www.cs.wisc.edu/condor/manual/v7.0/3_3Configuration.html#15357
and
http://www.cs.wisc.edu/condor/manual/v7.0/2_10DAGMan_Applications.html#S
ECTION003102900000000000000
Both those make it sound like DAGMan *tries* but can't actually limit
things to specifically N number of jobs.
Did I read that right?
Yes, DAGMan can only throttle things at the cluster level (because
throttling DAGMan just throttles the rate of condor_submits -- DAGMan
doesn't put jobs on hold or anything once they get into the queue).
If the original user can split their clusters up and just have one
job per cluster they could throttle things with DAGMan; otherwise DAGMan
won't help.
Kent Wenger
Condor Team