Differences

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

--- high_lod_tips [2016/12/15 17:06]
mgstauff created
+++ high_lod_tips [2017/10/27 18:38] (current)
mgstauff
@@ Line 1: / Line 1: @@
 When the cluster is very busy, your jobs might wait for much longer than usual in the queue before they run. Here are some tips about what you can do to help yourself and other users.
-) [[using_ogs_sge#using_basicq_queue_for_qsub_jobs|Use basic.q]]
+) [[using_ogs_sge#using_basicq_queue_for_qsub_jobs|Use basic.q]]. The older nodes in ''basic.q'' are usually very under-used.
-) Request less memory when you can.
+) [[using_ogs_sge#requesting_more_or_less_memory|Request less memory]] when you can.
 This is good for you and for the whole community.
-Right now, there are about 6 slots that could be used by jobs that request about 1.5GB or less vmem (the default request is 3GB)
-https://cfn.upenn.edu/computing/wiki/doku.php?id=using_ogs_sge#requesting_more_or_less_memory
+) [[using_ogs_sge#ending_an_interactive_session|Kill your qlogin sessions]] that you aren't using (to free resources for qsub jobs)
-) Kill your qlogin sessions that you aren't using (to free resources for others)
-https://cfn.upenn.edu/computing/wiki/doku.php?id=using_ogs_sge#ending_an_interactive_session

Computing