Help:HPCC Torque Queues: Difference between revisions
From CSDMS
m Add max run times to queue table |
m changed geshi to syntaxhighlight |
||
| Line 34: | Line 34: | ||
Use <tt>qstat</tt> to get an up-to-date summary of the queues. | Use <tt>qstat</tt> to get an up-to-date summary of the queues. | ||
< | <syntaxhighlight lang="bash" lines="0"> | ||
> qstat -q | > qstat -q | ||
Queue Memory CPU Time Walltime Node Run Que Lm State | Queue Memory CPU Time Walltime Node Run Que Lm State | ||
| Line 47: | Line 47: | ||
----- ----- | ----- ----- | ||
2 0 | 2 0 | ||
</ | </syntaxhighlight> | ||
Revision as of 18:41, 20 July 2009
Job Queues on the CSDMS HPCC
We have defined a set of queues to which you can submit your jobs to on beach. Note that if you do not specify a queue, it will be sent to the default queue. More importantly note that the default queue has a maximum run time of 4 days and so if your job has not finished after 4 days, it will be killed.
Cheat Sheet
A brief list of the job queues on the CSDMS HPCC. Please note that the queues are still being configured and that the following information may change without notice. We should get to a stable set of queues shortly.
| Queue | Max Run Time | Description |
|---|---|---|
| default | 4 days | Jobs that should be done within a few days |
| long | unlimited | Jobs that will run for a long time |
| debug | 2 hours | Short jobs that might be used for testing |
| fat | unlimited | Jobs that need to use memory-rich nodes |
| vip | 1 day | Important jobs that need to get done now |
Use qstat to get an up-to-date summary of the queues.
> qstat -q
Queue Memory CPU Time Walltime Node Run Que Lm State
---------------- ------ -------- -------- ---- --- --- -- -----
special_user -- -- -- -- 0 0 -- E R
workq -- -- -- -- 1 0 -- E R
debug -- 02:00:00 -- -- 0 0 -- E R
long -- -- -- -- 1 0 -- E R
fat -- -- -- -- 0 0 -- E R
vip -- 24:00:00 -- -- 0 0 -- E R
default -- 96:00:00 -- -- 0 0 -- E R
----- -----
2 0
