You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 3 Next »

This page contains information in no particular order. If a lot more information is added, one should think about organizing it.

SLURM auto-completion tool

For anyone using Slurm tool often, the following utilities is really helpful: https://github.com/SchedMD/slurm/tree/master/contribs/slurm_completion_help

sinfo

 sinfo is used to view partition and node information for a system running Slurm. 

Examples

sinfo -o "%C" -n sdfmilan[021-022,040,202-204,210-213,226,232]  
CPUS(A/I/O/T)
991/545/0/1536


( %C shows "allocated/idle/other/total") So 991 cores are still in use. With -o "%n %C"  one gets the usage per node:

sinfo -o "%n %C"  -n sdfmilan[021-022,040,202-204,210-213,226,232]
HOSTNAMES CPUS(A/I/O/T)
sdfmilan021 120/8/0/128
sdfmilan022 45/83/0/128
sdfmilan040 8/120/0/128
sdfmilan202 116/12/0/128
sdfmilan203 120/8/0/128
sdfmilan204 120/8/0/128
sdfmilan210 120/8/0/128
sdfmilan211 113/15/0/128
sdfmilan212 105/23/0/128
sdfmilan213 104/24/0/128
sdfmilan226 9/119/0/128
sdfmilan232 7/121/0/128

scontrol, sacctmgr

scontrol is used to view or modify Slurm configuration including: job, job step, node, partition, reservation, and overall system configuration. Most of the commands can only be executed by user root or an Administrator.

  • Detail job information: scontrol show jobid -dd <jobID>
  • Show reservation: scontrol show res

sacctmgr is used to deal with accounts, assocations and users (format can be modified at will. Remove it to see all):

  • Show what account a user is associated with: sacctmgr show associations Users=espov format=cluster,account,partition,QOS

Priorities

Show priorities for an account: sacctmgr list associations -p accounts=<accounts>

Show priority level for a job: sprio -j <jobID>

Show priority coefficients: sacctmgr show qos format=name,priority,usagefactor

  • No labels