sbatch

Use the ‘sbatch’ command to submit your jobs:

[johnchris@l001 ~]$ sbatch strata_test.sl
sbatch: slurm_job_submit: Exclusive/Full Node jobs are billed for all cores and memory for all requested nodes.
Submitted batch job 2217698

scancel

Use the ‘scancel’ command to withdraw a job:

[johnchris@l001 ~]$ scancel 2217697

sacct

‘sacct’ displays accounting data for all jobs and job steps in the Slurm job accounting log or Slurm Database, providing both live and historic data.

‘sacct’ can track users’ recent jobs and display job IDs and other details:

[johnchris@l001 ~]$ sacct

JobID          JobName     Partition   Account    AllocCPUS   State       ExitCode
------------   ---------   ----------  ---------  ----------  ----------  --------
2217697        sample      exclusive   johnchris   48          CANCELLED+  0:0
2217698        sample      exclusive   johnchris   48          PENDING     0:0
2217702        sample      exclusive   johnchris   48          PENDING     0:0
2217703        bash        shared      johnchris   4           RUNNING     0:0
2217703.ext+   extern                  johnchris   4           RUNNING     0:0
2217703.0      bash                    johnchris   4           RUNNING     0:0

‘sacct -j jobid’ will display only the specified job(s):

[johnchris@l001 ~]$ sacct -j 2193166

JobID          JobName     Partition   Account    AllocCPUS   State       ExitCode
------------   ---------   ----------  ---------  ----------  ----------  --------
2216924        sample      shared      johnchris   12          RUNNING     0:0
2216924.bat+   batch       shared      johnchris   12          RUNNING     0:0
2216924.ext+   extern      shared      johnchris   12          RUNNING     0:0

‘sacct’ can provide a detailed report on the memory, CPU, and other usage metrics of a job:

[johnchris@l001 ~]$ sacct -j 2412831 --format JobID,Elapsed,ReqMem,MaxRSS,AllocCPUs,TotalCPU,State

JobID          Elapsed    ReqMem   MaxRSS  AllocCPUS  TotalCPU  State
------------   --------   -------  -------  ---------  --------  ---------
2412831        03:39:19   128Gn    128      128        00:00:00  RUNNING
2412831.bat+   03:39:19   128Gn    32       128        00:00:00  RUNNING
2412831.ext+   03:39:19   128Gn    128      128        00:00:00  RUNNING
2412831.0      03:38:50   128Gn    3        128        00:00:00  RUNNING

scontrol

‘scontrol show job jobid’ will show detailed information about the job with the specified ID, including the job script:

[johnchris@l001 ~]$ scontrol show job 2412831

JobId=2412831 JobName=amber_md
   UserId=johnchris(12345678) GroupId=johnchris(12345678) MCS_label=N/A
   Priority=24310 Nice=0 Account=drjohn QOS=normal
   JobState=RUNNING Reason=None Dependency=(null)
   Requeue=1 Restarts=0 BatchFlag=1 Reboot=0 ExitCode=0:0
   RunTime=03:55:29 TimeLimit=3-00:00:00 TimeMin=N/A
   SubmitTime=2024-06-10T20:43:05 EligibleTime=2024-06-10T20:43:05
   AccrueTime=2024-06-10T20:43:05
   StartTime=2024-06-10T20:43:13 EndTime=2024-06-13T20:43:13 Deadline=N/A
   SuspendTime=None SecsPreSuspend=0 LastSchedEval=2024-06-10T20:43:13
   Partition=shared AllocNode:Sid=l001:2156446
   ReqNodeList=(null) ExcNodeList=g[001-012]
   NodeList=c[013-015,027]
   BatchHost=c013
   NumNodes=4 NumCPUs=128 NumTasks=128 CPUs/Task=N/A ReqB:S:C:T=0:0:*:*:*
   TRES=cpu=128,mem=512G,node=4,billing=256
   Socks/Node=* NtasksPerN:B:S:C=32:0:*:* CoreSpec=*
   MinCPUsNode=32 MinMemoryNode=128G MinTmpDiskNode=0
   Features=(null) DelayBoot=00:00:00
   OverSubscribe=OK Contiguous=0 Licenses=(null) Network=(null)
   Command=/mmfs1/data/johnchris/sample/test/example.slm
   WorkDir=/mmfs1/data/johnchris/sample/test
   StdErr=/mmfs1/data/johnchris/sample/test/slurm-2412831.out
   StdIn=/dev/null
   StdOut=/mmfs1/data/johnchris/sample/test/slurm-2412831.out
   Power=

Scroll to Top