The following information is largely replicated from SchedMD’s salloc man page, and is the subset that is likely most relevant to most NCCS users.
The salloc command honors the following environment variables, when present (these override any inline directives within your batch script, but will be overridden by those also specified on the salloc command line):
Input Variables
| SALLOC_ACCOUNT | Same as -A, –account |
| SALLOC_BELL | Same as –bell |
| SALLOC_IMMEDIATE | Same as -I, –immediate |
| SALLOC_JOB_NAME | Same as -J, –job-name |
| SALLOC_QOS | Same as -q, –qos |
| SALLOC_TIMELIMIT | Same as -t, –time |
and sets the following output variables, which are present in the job’s environment during execution (run ‘env | grep SLURM’ within your job, to obtain a full list):
Output Variables
| SLURM_JOB_ID | Numeric ID of the job allocation |
| SLURM_JOB_CPUS_PER_NODE | Count of processors available to the job on this node. |
| SLURM_CPUS_PER_TASK | Number of cpus requested per task. Only set if the –cpus-per-task option is specified. |
| SLURM_DISTRIBUTION | Same as -m, –distribution |
| SLURM_JOB_NODELIST | (and SLURM_NODELIST for backwards compatibility): List of nodes allocated to the job. |
| SLURM_JOB_NUM_NODES | (and SLURM_NNODES for backwards compatibility): Total number of nodes in the job’s resource allocation. |
| SLURM_JOB_PARTITION | Name of the partition in which the job is running. |
| SLURMD_NODENAME | Name of the node upon which the current job step executes (value is equal to output of linux ‘hostname -s’ command). |
| SLURM_NTASKS (and SLURM_NPROCS for backwards compatibility) | Same as -n, –ntasks |
| SLURM_NTASKS_PER_NODE | Set to the value of –ntasks-per-node, if specified |
| SLURM_PROCID | The MPI rank (or relative process ID) of the current process |
| SLURM_RESTART_COUNT | If the job has been restarted due to system failure or has been explicitly requeued, this will be set to the number of times the job has been restarted. |
| SLURM_SUBMIT_DIR | The directory from which salloc was invoked. |
| SLURM_SUBMIT_HOST | The host from which salloc was invoked. |
| SLURM_TASKS_PER_NODE | Number of tasks to be initiated on each node. Values are comma separated and in the same order as SLURM_NODELIST. If two or more consecutive nodes are to have the same task count, that count is followed by “(x#)” where “#” is the repetition count. For example, “SLURM_TASKS_PER_NODE=2(x3),1” indicates that the first three nodes will each execute three tasks and the fourth node will execute one task. |
For more information on salloc, please refer to the man pages. Use the following command, after you’ve logged onto Discover:
man sallocor
salloc -help