LSF to SLURM quick reference
This article shows the SLURM equivalent of LSF commands. Note: The recording and PPT presentation of the first webinar on transitioning from LSF to SLURM are now available at https://www.ceda.ac.uk/events/transitioning-to-slurm-webinar
The SLURM Scheduler
SLURM (formerly known as Simple Linux Utility for Resource Management or SLURM) is the job scheduler deployed on JASMIN. It allows users to submit, monitor, and control jobs on the CentOS7 LOTUS sub-cluster.
Table 1 Essential LSF/SLURM commands
| LSF | SLURM | Description |
| bsub < script_file | sbatch script_file | Submit a job script to the scheduler |
| bqueues | sinfo | Show available scheduling queues |
| bjobs | squeue | List user's pending and running jobs |
Table 2 Job specification
| LSF | SLURM | Description |
| #BSUB | #SBATCH | Scheduler directive |
| -q queue_name | -p queue_name |
Specify the scheduling queue |
| -W hh:mm:ss | --time= hh:mm:ss or -t hh:mm:ss |
Set the maximum runtime limit |
| -We hh:mm:ss | --time-min= hh:mm:ss |
Set an estimated runtime |
| -J job_name | --job-name= jobname |
Specify a name for the job |
| -o filename, -e filename -oo/-eo filename |
--output= filename or -o filename, --error=filename or -e filename The default file name is "slurm-%j.out", where the "%j"is replaced by the job ID For job arrays, the default file name is "slurm-%A_%a.out", "%A" is replaced by the job ID and "%a" with the array index. --open-mode=append|truncate |
Standard job output and error output. Default append. Overwrite job error/output files |
| %J %I |
%j %a |
Job ID for -oo/eo filename Job array index |
| -R "rusage[mem= XXX]" |
--mem= XXX | Memory XXX required for the job. Default units are megabytes |
| -J job_name[index_list] | --array= index (e.g. --array=1-10) |
specify a job array |
| -J job_name[index_list]%number-of-simultaneous-jobs | --array=index%ArrayTaskThrottle e.g. --array=1-15%4 will limit the number of simultaneously running tasks from this job array to 4 |
A maximum number of simultaneously running tasks from the job array may be specified using a "%" separator. |
| -cwd directory | -D, --chdir=<directory> | |
| -x | --exclusive | Exclusive execution mode |
| -P project | -A account-name or --account=account-name Note: The account has to be defined by the SLURM administrator and the user assigned to the account. |
Charge resources used by this job to the specified account or project |
| -n number-of-cores | --ntasks= number-of-cores or -n number-of-cores | Number of CPU cores |
| -m <host-group-name> | --constraint="<host-group-name>" | To select a node with a specific processor model and memory, |
| |
Table 3 Job control commands
| LSF | SLURM | Description |
| bkill jobid | scancel jobid | Kill a job |
| bjobs -l jobid | scontrol show job jobid | Show details job
|
| bmod jobid | scontrol update job jobid | Modify a pending job |
| |
||
| |
Table 4 Job environment variables
| LSF | SLURM | Description |
| $LSB_JOBID | $SLURM_JOBID | Job identifier number |
| $LSB_JOBID |
$SLURM_ARRAY_JOB_ID | Job array |
| $LSB_JOBINDEX | $SLURM_ARRAY_TASK_ID | Job array index |
| $LSB_JOBINDEX_END | $SLURM_ARRAY_TASK_MAX | Last index number within a job array |
| $LSB_MAX_NUM_PROCESSORS | $SLURM_NTASKS | Number of processors allocated |
| |