...
Please note the additional parameters and the MPI-enabled application! This job uses 2 nodes, with 48 tasks on each node, and each task uses 1 CPU. Each node participating in execution will allocate 184GB of memory. The job's stdout and stderr are redirected to respective joblog.txt and joberr.txt files. Note that existing output files will be overwritten. This example assumes that In example the myapp.bin application is an uses MPI application , and that the mpiexec command is responsible for spawning the additional application ranks (processes). In most cases, mpiexec handles the application configuration (e.g., number of processes) by communicating with Slurm, so explicitly specifying the -np parameter is not required.
...
| Code Block | ||||
|---|---|---|---|---|
| ||||
#!/bin/bash #SBATCH --job-name=job_name #SBATCH --time=01:00:00 #SBATCH --account=grantname-gpu #SBATCH --partition=plgrid-gpu-v100 #SBATCH --cpus-per-task=4 #SBATCH --mem=40G #SBATCH --gres=gpu module load cuda ./myapp |
...