User Tools

Site Tools



This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
Last revision Both sides next revision
condor [2015/09/30 17:43]
sertalpbilal Some improvements in CONDOR commands and sample case
condor [2016/10/07 09:25]
Line 1: Line 1:
 ======CONDOR====== ======CONDOR======
 +<note warning>​If you are a new user, please use [[tutorial:​torque|Torque]]</​note>​
 +<note important>​CONDOR will be deprecated. You need to switch new scheduling system [[tutorial:​torque|Torque]] at your earliest chance.</​note>​
 ===== What is CONDOR ===== ===== What is CONDOR =====
Line 261: Line 265:
 ===== Some other CONDOR commands ===== ===== Some other CONDOR commands =====
-^ Command ^ Action ^+^ Command ^ Action ​^ Info ^
 | ''​condor_userprio''​ | shows the user priority | condor_userprio | | ''​condor_userprio''​ | shows the user priority | condor_userprio |
 | ''​condor_status''​ | show the current status of CONDOR nodes | | ''​condor_status''​ | show the current status of CONDOR nodes |
Line 269: Line 273:
 FIXME To submit MPI jobs to our condor pool you can check Dr. Takac'​s [[http://​​mpi|MPI tutorial]] FIXME To submit MPI jobs to our condor pool you can check Dr. Takac'​s [[http://​​mpi|MPI tutorial]]
 +===== Using AMPL with Condor =====
 +We have limited license of AMPL installed in COR@L Lab. The license allows at most 10 simultaneous AMPL jobs. If you are using AMPL in your experiments you can let condor know about this and it will schedule all jobs that needs AMPL considering the license limit. For this you should add the following line to your condor submission file.
 +''​concurrency_limits = AMPL''​
 +===== Condor Jobs Memory Usage =====
 +Please check status of your condor jobs regularly, especially memory usage.
 +Each polyp node has 16 processors and 32 GB memory. This means 1 process
 +gets 2 GB memory in average.
 +When a polyp node is out of memory it starts using hard drive (swap) as memory
 +but reading and writing from hard drives is 1000 times slower. This means
 +if your jobs are using large amounts of memory and the polyp node processing
 +your job is out of memory, do not expect your job to terminate.
 +You can see memory usage of your job using ''​condor_q''​ command (7th column
 +gives memory usage in MB).
 +You can check the node your job is running using ''​condor_q -run''​
 +You can check memory status in a node using ''​ssh polyp6 "​vmstat -s"''​.
 +For more memory checking commands see http://​​linux-command-check-memory-usage/​
 +or google is your friend.
 +**Your job might get killed if it is using swap. Do not waste your
 +system administrators'​ time with this and force them to police the condor jobs.
 +Just control your jobs and submit jobs that are reasonable.**
condor.txt ยท Last modified: 2017/07/16 00:05 by sertalpbilal