Accessing the high memory nodes (fat nodes on lonsdale)
There are a number of large memory nodes on lonsdale which can be accessed if the users jobs require these nodes to do so additional specifications must be given to srun, salloc or sbatch.
For example to access the 32GB nodes using a batch script:
#!/bin/sh #SBATCH -N 1 # 1 node #SBATCH -t 03:00:00 # 1 day and 3 hours #SBATCH -p compute # parition name #SBATCH -U myaccount # your project name - contact Ops if unsure what this is #SBATCH -J my_job_name # sensible name for the job #SBATCH --mem=32194 # this is specified in megabytes per node srun hostname
Or srun:
$ srun --mem=32194 -N 1 -p compute -t 03:00:00 -U myaccount hostname lonsdale-n013.cluster lonsdale-n014.cluster
Or salloc:
$ salloc --mem=32194 -N 1 -p compute -t 03:00:00 -U myaccount