BlueBEAR Large Memory Service

Each of BlueBEAR's Icelake nodes (introduced autumn 2021) have 499G available memory (RAM) and are therefore suitable for tasks that require a larger amounts of memory. The simplest method for requesting a large amount of memory is to include the following #SBATCH headers in your job script:

#SBATCH --nodes 1
#SBATCH --mem 248G

Efficient Use

In order to ensure that BlueBEAR's resources are used efficiently our preference is that resource requests work on the basis of quarter-nodes slices, which means that other jobs can run on the same node, concurrently with a larger-memory job. Please therefore request --mem in multiples of 124G.

Another solution is to use the --mem-per-cpu Slurm option, and then scale using --ntasks. For example:

#SBATCH --nodes 1
#SBATCH --mem-per-cpu 6900M
#SBATCH --ntasks 54

In the above example we're specifying 54 cores (or three quarters of the available 72 cores) in order to get 372600M memory.

BlueBEAR Large Memory Service (a.k.a. bblargemem QOS)

Due to the changes in BlueBEAR's topology as outlined above, the bblargemem QOS has been retired as the higher-memory Icelake nodes are available via the standard QOSes.

Colleges

Professional Services