Difference: BatchSystem (12 vs. 13)

Revision 132016-04-22 - GordonStewart

Line: 1 to 1

META TOPICPARENT	name="WebHome"

Batch System

Changed:

<
< The PPE group maintains a PBS cluster for running small quantities of jobs. If you need to run large numbers of jobs, you should investigate the possibility of running on ScotGrid.

>
>

Batch System

Changed:

<
<

The batch system uses the TORQUE

resource manager (based on OpenPBS) and the Maui scheduler. It can be accessed from any Linux desktop using the commands described below.

The current composition of the batch system is as follows:

>
> The PPE group maintains a PBS cluster for running small quantities of jobs. If you need to run large numbers of jobs, you should investigate the possibility of running on ScotGrid. The current composition of the batch system is as follows:

Nodes	Operating System	Total CPU Cores

Changed:

<
<

node123 to node456 SL5 999

>
>

`node007`	Scientific Linux 6	40
`node008`	Scientific Linux 5	4
`node013` to `node017`	Scientific Linux 5	20
`node019`	Scientific Linux 6	4
`node034`	Scientific Linux 6	56
`tempnode001` to `tempnode006`	Scientific Linux 5	24
`tempnode007` to `tempnode015`	Scientific Linux 6	36

The following queues are provided:

Name	Operating System	Maximum runtime

Changed:

<
<

`short5`	SL5	1 hour
`medium5`	SL5	6 hours
`long5`	SL5	1 day
`vlong5`	SL5	5 days
`short6`	SL6	1 hour
`medium6`	SL6	6 hours
`long6`	SL6	1 day
`vlong6`	SL6	5 days

>
>

`short5`	Scientific Linux 5	1 hour
`medium5`	Scientific Linux 5	6 hours
`long5`	Scientific Linux 5	1 day
`vlong5`	Scientific Linux 5	5 days
`short6`	Scientific Linux 6	1 hour
`medium6`	Scientific Linux 6	6 hours
`long6`	Scientific Linux 6	1 day
`vlong6`	Scientific Linux 6	5 days

Jobs running in the vlong* queues can be pre-empted by jobs in the short* and medium* queues. A pre-empted job is placed in the suspended state; it remains in memory on the compute node, but is no longer being executed. Once the pre-empting job has finished, the pre-empted job will be allowed to continue.

The PBS headnode is offler.ppe.gla.ac.uk, and you will see this name in the output of various PBS commands.

Using PBS

Added:

>
> Batch jobs can be submitted and managed from any Linux desktop using the commands described in this section. Further information on these commands can be found in the linked documentation and Linux man pages at the bottom of this page.

Create a submission script

Jobs are defined using a submission script, which is like a shell script with the addition of certain directives (indicated by the #PBS prefix) which tell PBS how the job should be handled. A simple submission script might look like the following:

Line: 44 to 55

$ qsub <FILENAME>

Changed:

<
< To submit a job defined by the submission script test.pbs:

>
> After running this command, the ID of the newly-submitted will be output. For example, to submit a job defined by the submission script test.pbs:

Changed:

<
< $ qsub test.pbs

>
>

$ qsub test.pbs
1000150.offler.ppe.gla.ac.uk

Changed:

<
< More details can be found in the qsub man page.

>
> The numerical portion of this ID (1000150 in this example) can be used to manage the job in the future.

Show running jobs

Added:

>
>

You can view details of submitted jobs using the qstat command:

$ qstat

offler.ppe.gla.ac.uk:
                                                                         Req'd  Req'd   Elap
Job ID               Username Queue    Jobname          SessID NDS   TSK Memory Time  S Time
-------------------- -------- -------- ---------------- ------ ----- --- ------ ----- - -----
1000151.offler.p     rrabbit  medium6  maus_sim_814      56289     1   1    --  05:59 R 03:21   node034
1000152.offler.p     bbunny   long6    test_job          29669     1   1    --  24:00 R 01:24   node007

This

Line: 78 to 106

To see the current priorities of waiting jobs use the command showq -i.

Deleted:

<
<

Job Pre-emption

Jobs in the vlong4 and vlong5 queues can be preempted by jobs waiting in the short4, short5, medium4 or medium5 queues. A preempted job is placed in the suspended state - it remains in memory but is not longer being executed. Once the preempting job has finished the preempted job starts executing again.

Job Priority

The priority of a job is the sum of several weighting factors.

Line: 90 to 114

The primary weighting is user fairshare. As a users jobs run their usage increases and the priority of their queued jobs decreases. This is balanced so that a user who uses exactly their fairshare allotment (currently 20% of the cpu averaged over the previous 48 days) will have their medium job priority decreased such that the medium job priority is equal to someone else's vlong job priority who has not used the batch system in the previous 48 days.
Waiting jobs priority slowly increases as a function of time waiting in the queue. Currently a vlong job would have to wait several weeks to match the priority of a medium queue job all other things being equal.

Changed:

<
<

Killing a job

>
>

Delete a job

Jobs are deleted using the qdel command:

$ qsub <JOB_ID>

To delete the job with ID 12345:

$ qdel 12345

References

Changed:

<
< Jobs may be terminated by executing qdel JOBID where the JOBID is the numerical ID code returned in the qstat listing.

>
>

qsub
qdel

View topic | History: r16 < r15 < r14 < r13 | More topic actions...