Slurm Partitions (Compute Resources)

In Slurm, jobs are submitted to partitions – think of these as job queues for different types of jobs.

In your jobscript, you must specify the partition, number of cores and max “wallclock” time your job is allowed to run for (up to a permitted maximum time.) There is now no default wallclock time – you must specify it.

You may also need to specify other flags depending on which parition you submit to. For example, if using one of the GPU partitions, you’ll need to say how many GPUs you want. The HPC Pool and H200 partitions also require an account code.

The table below provides an overview of the partitions, and the compute resources, currently available in the upgraded CSF3 cluster (Slurm).

The Slurm jobscript template (shown further down this page) gives you a starting point for your jobscript. Much more detail on Slurm jobscript options is available in the sbatch command page and, for users of previous CSF systems, the SGE to Slurm reference.

You will also need to use Slurm commands on the login node: sbatch, squeue, scancel and srun.

Reminder: You must specify a Slurm Partition in your jobscript for all job types: -p nameYou must also specify a maximum wallclock time limit for your job (up to a permitted maximum time): -t timelimit

See table below.

CSF3 Slurm Partitions and Compute Resources
Slurm Partition name	Partition Summary	CPU	Max Wallclock	Compute Resources	Notes
Parallel Jobs
`-p multicore_small`	Single-node parallel (2-40 cores) batch jobs. `-n numcores`	Intel	7 days `-t 7-0`	21x Haswell nodes (24 cores, 4-5GB/core) Optional: `-C haswell` 28x Skylake nodes (32 cores, 4-5GB/core) Optional: `-C skylake` NEW: 70x Cascadelake nodes (40 cores, 4GB/core) Optional: `-C cascadelake` or `-C csf4`	1400 Haswell and Skylake cores available (shared with the serial partition below.) 2800 Cascadelake cores now available.. Unless specified, 4GB/core will be allocated no matter which node-type your job runs on. Recommended for jobs of 2 to 40 cores. multicore (see below) is suggested for jobs of 41 to 168 cores.
`-p multicore`	Single-node parallel (2-168 cores) batch jobs. `-n numcores`	AMD Genoa	7 days `-t 7-0`	84x AMD nodes (168 cores per node). 8GB RAM per core.	14,112 cores available in total. Recommended for jobs of 41 to 168 cores. multicore_small (see above) is suggested for jobs of 2 to 40 cores.
`-p interactive`	Serial (1-core) and small parallel interactive jobs. Optional: `-n numcores` Serial (1-core) batch jobs.	AMD Genoa	6 hours `-t 0-6`	2x AMD nodes (168 cores per node). 8GB RAM per core.	336 cores available in total. `srun` (interactive), `srun-x11` (interactive GUIs) and also `sbatch` (jobscripts) accepted. For testing apps on AMD “Genoa”.
Serial Jobs
`-p serial`	Serial (1-core) batch jobs. Serial (1-core) interactive jobs.	Intel	7 days `-t 7-0` 6 hours `-t 0-6`	21x Haswell nodes (24 cores, 5GB/core) Optional: `-C haswell` 28x Skylake nodes (32 cores, 6GB/core) Optional: `-C skylake`	1400 cores available in total (shared with the multicore_small partition above)
High Memory Jobs
`-p himem`	High memory jobs up to 2TB (1-32 cores). Optional: `-n numcores`	Intel	7 days `-t 7-0`	1x 512GB node (16 cores, 32GB per core) Optional: `-C haswell` 4x 1.5TB node (32 cores, 40GB per core) Optional: `-C cascadelake` 9x 2TB nodes (32 cores, 64GB per core) Optional: `-C icelake` or `-C ssd`	Job default is 32GB/core unless you ask for more – even if you add a `-C` constraint! Use `--mem=totalG` or `--mem-per-cpu=numG` to request more. High Memory Docs
`-p vhimem`	High memory jobs requiring up to 4TB. Optional: `-n numcores`	Intel	7 days `-t 7-0`	1x 4tb node (32 cores, 128GB per core) Optional: `-C icelake` or `-C ssd`	Restricted access – request access
GPU Jobs
`-p gpuA`	Nvidia A100 GPU batch and interactive jobs. Optional: `-n numcores` Required: `-G NUM` or `--gpus=NUM`	AMD Milan	4 days `-t 4-0`	Node spec: 4x Nvidia A100s 80GB GPU RAM 48x AMD “Milan” host cores 512GB host RAM (10GB/core)	Max 12 CPU-cores per GPU 19 nodes available = 76 GPUs
`-p gpuL`	Nvidia L40S GPU batch and interactive jobs. Optional: `-n numcores` Required: `-G NUM` or `--gpus=NUM`	Intel Xeon Sapphire Rapids	4 days `-t 4-0`	Node spec: 4x Nvidia L40S 48GB GPU RAM 2 x 24-core Intel “Sapphire Rapids” host cores 512GB host RAM (10GB/core)	Max 12 CPU-cores per GPU 22 nodes available = 88 GPUs
`-p gpuA40GB`	Nvidia A100 (40GB) GPU batch and interactive jobs. Optional: `-n numcores` Required: `-G NUM` or `--gpus=NUM`	AMD Milan	4 days `-t 4-0`	Node spec: 4x Nvidia A100s 40GB GPU RAM 48x AMD “Milan” host cores 512GB host RAM (10GB/core)	Very Restricted Access Max 12 CPU-cores per GPU, 2 nodes available = 8 GPUs
`-p gpuH`	Nvidia H200 (141 GB) GPU batch jobs. Optional: `-n numcores` or `-c numcores` Required: `-A accountcode` `-G NUM` or `--gpus=NUM`	Intel Xeon Emerald Rapids	4 days `-t 4-0`	Node spec: 8x Nvidia H200 141GB GPU RAM 64 “Emerald Rapids” host cores 1.5 TB host RAM (24GB/core)	Restricted Access Max 8 CPU-cores per GPU, up to 3 nodes available = 24 GPUs. H200 docs.
`-p gpuH_short`	Nvidia H200 (141 GB) GPU batch and interactive jobs. Optional: `-n numcores` or `-c numcores` Required: `-A accountcode` `-G NUM` or `--gpus=NUM`	Intel Xeon Emerald Rapids	1 day `-t 1-0`	Node spec: 8x Nvidia H200 141GB GPU RAM 64 “Emerald Rapids” host cores 1.5 TB host RAM (24GB/core)	Restricted Access Max 8 CPU-cores per GPU, up to 2 nodes available = 16 GPUs. H200 docs.
HPC Pool
`-p hpcpool`	Multi-node parallel (120-1000 cores) batch jobs. Required: -A hpc-proj-code -N numnodes -n numcores	Intel Cascadelake	4 days `-t 4-0`	Cascadelake nodes (40 cores, 4GB/core)	Restricted Access 2000 cores available in total. HPC Pool docs.
Multicore (coming soon)

*Information is correct as of 1st October 2026

Jobscript Template

All jobs must specify the partition name (see above) and wallclock runtime limit (see the various time formats) for the job.

Note that the Slurm lines begin with #SBATCH – not #$ as used on SGE. A common mistake is to use a $ symbol in the Slurm special line. It is NOT #$BATCH.

CPU Jobs

#!/bin/bash --login
#SBATCH -p partitionname    # Required (all jobs) - see table above
#SBATCH -n numcores         # Required (parallel jobs)  - defaults to 1 for serial
#SBATCH -t timelimit        # Required (all jobs) wallclock timelimit. Max permitted is 7 days (7-0)

#SBATCH flag resources      # See table above for additional high-memory, CPU architecture
                            # and GPU resource flags.

# We recommend purging your env
module purge
module load apps/gcc/someapp/1.2.3

# $SLURM_NTASKS will be set to the number of cores requested above, if your app wants to know.
someapp.exe

# Slurm knows to run $SLURM_NTASKS mpi processes if using mpirun
mpirun some_mpi_app.exe

For OpenMP parallel applications you can use:

#!/bin/bash --login
#SBATCH -n 1           # 1 task ($SLURM_NTASKS set to 1)
#SBATCH -c numcores    # cores-per-task ($SLURM_CPUS_PER_TASK set to this)
#SBATCH -t timelimit   # Max wallclock time

# Inform OpenMP how many cores to use
export OMP_NUM_THREADS=$SLURM_CPUS_PER_TASK
some_omp_app.exe

Submit the job using

sbatch jobscript

Check on the status of the job using

squeue

GPU Jobs

Please see the GPU Jobs page.

High-memory Jobs

Please see the High Memory Jobs page.

More jobscript options

For much more information on Slurm jobscript options, compared to SGE options, please see the SGE to Slurm reference.

Last modified on October 6, 2026 at 5:29 pm by George Leaver