Skip to end of metadata
Go to start of metadata
The program pbstop, available on the login nodes, shows which jobs are currently running on which nodes and cores of a cluster.

Jobs belonging to a single user can be highlighted by launching pbstop with the -u switch:

pbstop -u <NetID>

(of course, replace <NetID> with your NYU NetID). Or, you can use the alias "me":

pbstop -u me

When you start pbstop you see something like the annotated screenshot below. You might need to resize your terminal to make it all fit: 

Each processor in each node is shown, so for example we can see here that the nodes in the first block have only 8 cores each, while those in the second block have 12.

  • Just below the center of the output, two nodes are shown separately due to having significantly different processor counts

This is good for seeing generally how busy the cluster is, but it is more useful to highlight only your own jobs. The screenshot below shows the jobs of one (arbitrarily selected) user:

The symbols used in this map are:

  • A period (.) indicates an idle processor (ie, CPU core)
  • A percentage symbol (%) indicates a core which is offline.
  • An ampersand (@) means a core allocated to a job not belonging to the selected user
  • Each other, colored symbol shows cores used by the same job. The bottom of the diagram has a key and the first several jobs

 

 

 

  • No labels