Difference between revisions of "Slurm"
From Systems Group
Systemswiki (Talk | contribs) (→SLURM Directives) |
Systemswiki (Talk | contribs) (→SLURM Directives) |
||
| Line 9: | Line 9: | ||
! Resource !! Syntax !! Example !!Description | ! Resource !! Syntax !! Example !!Description | ||
|- | |- | ||
| − | | Account || --account=slurmgeneral || --account=slurmgeneral || entity which resources are charged to. | + | | Account || --account=slurmgeneral || --account=slurmgeneral || entity which resources are charged to. [https://systems.cs.odu.edu/Slurm#Compute_Resources|available accounts] |
|- | |- | ||
| − | | Partition || --partition=slurm-general-01 || --partition=slurm-general-01 || where job resources are allocated. | + | | Partition || --partition=slurm-general-01 || --partition=slurm-general-01 || where job resources are allocated. [https://systems.cs.odu.edu/Slurm#Compute_Resources|available partitions] |
|- | |- | ||
| Job Name || --job-name=<filename> || --job-name=testprogram || name of job to be queued | | Job Name || --job-name=<filename> || --job-name=testprogram || name of job to be queued | ||
Revision as of 19:31, 4 August 2022
Slurm is an open-source job scheduler for Linux and Unix-like kernels.
SLURM Directives
SLURM directives are job options that constrain the job to the conditions specified. Directives can be identified by the syntax `#SBATCH <flag>'. Commonly used flags are listed below.
| Resource | Syntax | Example | Description |
|---|---|---|---|
| Account | --account=slurmgeneral | --account=slurmgeneral | available accounts] |
| Partition | --partition=slurm-general-01 | --partition=slurm-general-01 | available partitions] |
| Job Name | --job-name=<filename> | --job-name=testprogram | name of job to be queued |
| Task | --ntask=<number> | --ntask=2 | useful for commands to be ran in parallel |
| Memory | --mem=<size>[units] | --mem=1gb | memory to be allocated for job |
| Output | --output=<filename> | --output=testprogram.log | name of job output file |
| Time | --time= | --time=01:00:00 | time limit for job |
SRUN
srun is used to submit jobs for execution in real time. Also used to create job steps.
srun example
srun --partition slurm-general-01 --account=slurmgeneral --pty /bin/bash # shell on compute node
# specifying which partition and account (applicable if assigned multiple accounts)
srun --pty /bin/bash # shell on compute node
# default partition and account is used when not specified
# This is not suggested as submitted jobs can be put in a pending stay due to incorrect permissions
SBATCH
Sbatch is a command used to submit jobs via batch scripts to SLURM.
batch script example
#!/bin/bash -l # login shell (required for lmod) #SBATCH --job-name=testprogram # job name #SBATCH --partition=slurm-general-01 # specifying which partition to run job on #SBATCH --account=slurmgeneral # only applicable if user is assigned multiple accounts #SBATCH --ntasks=1 # commands to run in parallel #SBATCH --mem=1gb # request 1gb of memory #SBATCH --output=testprogram.lob # output and error log date sleep 10 module use /mnt/lmod_modules/Linux/ module load miniconda3 someProgram.py date
submitting a job using sbatch
sbatch myprogram.sh # queue job using a batch script
sbatch --partition slurm-general-01 --account=slurmgeneral myprogram.sh # batch script specifying which partition and account when not specified using a slurm directive within the script
Compute Resources
The ODU CS department HPC cluster is comprised of multiple partitions where users can submit jobs. Each partition can only be accessed by users who are assigned to the partitions respective account. Not all partitions can be accessed by all users.
| Cluster | Partition | Account |
|---|---|---|
| slurm-cluster | slurm-general-01 | slurmgeneral |
| slurm-cluster | slurm-general-02 | slurmgeneral |
| slurm-cluster | haoresearch | shaoresearch |
| slurm-cluster | lusiliresearch | lliresearch |
| slurm-cluster | wangresearch | fwangresearch |
Troubleshooting
How to view assigned account
sacctmgr show association -p user=$username
