How can I tell how much memory my slurm job used or needs?

Locales: Data Centers and Campus RC Yale
1

I don’t want to run out of RAM, but I also want to schedule my jobs efficiently.

2

There are a number of tools available to monitor the resource usage of your jobs. Our documentation page has a detailed discussion (link).

For completed jobs, the seff tool reports the memory and CPU utilization:

$ seff 21294645
Job ID: 21294645
Cluster: farnam
User/Group: rdb9/lsprog
State: COMPLETED (exit code 0)
Cores: 1
CPU Utilized: 00:15:55
CPU Efficiency: 17.04% of 01:33:23 core-walltime
Job Wall-clock time: 01:33:23
Memory Utilized: 446.20 MB
Memory Efficiency: 8.71% of 5.00 GB

For running jobs, you are able to log into the compute node where the job is running (eg, c01n05) and inspect the job with top:

[rdb9@farnam1 ~]$ ssh c01n05
[rdb9@c01n05 ~]$ top

The column titled “RES” will list the amount of memory your processes are using.