Loading src/hpc_guide/slurm.md +15 −2 Original line number Diff line number Diff line Loading @@ -4,6 +4,18 @@ order: 1 # SLURM guide ## Inspect running job You can further inspect a running job by connecting to it with this command : ```sh srun --jobid=jobid --overlap --pty bash ``` This will open an interactive shell as a job step under an already allocated job. I.e. you will be able to see how your job is behaving. For distributed memory jobs you will get a shell at the first node used by your job. ## Debug in Real-time on SLURM ### Problem Loading @@ -15,7 +27,7 @@ process. It goes something like this : 1. Check for errors/change code. 1. Repeat endlessly until your code works. ### Solution ### Solution : interactive jobs (`srun --pty`) Fortunately, there’s a better way, you can debug in real-time like so: Loading @@ -35,6 +47,7 @@ Fortunately, there’s a better way, you can debug in real-time like so: 1. Check for errors/change code continuously until code is fixed or node has timed out. ### Sources ## Sources - [HPC-UiT Services User Documentation](https://hpc-uit.readthedocs.io/en/latest/jobs/interactive.html#starting-an-interactive-job) - [Innsbruck University's SLURM Tutorial](https://www.uibk.ac.at/zid/systeme/hpc-systeme/common/tutorials/slurm-tutorial.html) Loading
src/hpc_guide/slurm.md +15 −2 Original line number Diff line number Diff line Loading @@ -4,6 +4,18 @@ order: 1 # SLURM guide ## Inspect running job You can further inspect a running job by connecting to it with this command : ```sh srun --jobid=jobid --overlap --pty bash ``` This will open an interactive shell as a job step under an already allocated job. I.e. you will be able to see how your job is behaving. For distributed memory jobs you will get a shell at the first node used by your job. ## Debug in Real-time on SLURM ### Problem Loading @@ -15,7 +27,7 @@ process. It goes something like this : 1. Check for errors/change code. 1. Repeat endlessly until your code works. ### Solution ### Solution : interactive jobs (`srun --pty`) Fortunately, there’s a better way, you can debug in real-time like so: Loading @@ -35,6 +47,7 @@ Fortunately, there’s a better way, you can debug in real-time like so: 1. Check for errors/change code continuously until code is fixed or node has timed out. ### Sources ## Sources - [HPC-UiT Services User Documentation](https://hpc-uit.readthedocs.io/en/latest/jobs/interactive.html#starting-an-interactive-job) - [Innsbruck University's SLURM Tutorial](https://www.uibk.ac.at/zid/systeme/hpc-systeme/common/tutorials/slurm-tutorial.html)
