Nimbulus Compute Documentation https://nimbulus.xyz/dokuwiki/ 2026-04-19T03:11:57+00:00 Nimbulus Compute Documentation https://nimbulus.xyz/dokuwiki/ https://nimbulus.xyz/dokuwiki/lib/exe/fetch.php?media=wiki:dokuwiki.svg text/html 2025-04-26T04:23:07+00:00 Anonymous (anonymous@undisclosed.example.com) cvmfs https://nimbulus.xyz/dokuwiki/doku.php?id=cvmfs&rev=1745641387&do=diff CVMFS (CERN Virtual Machine File System (CVMFS) is the FUSE filesystem that computecanada uses to distribute software. They also allow non-computecanada clusters to use their distributions, and that is exactly what we are doing. Installation (Already done on systems, if need to redo it): text/html 2025-04-26T04:23:07+00:00 Anonymous (anonymous@undisclosed.example.com) gcc https://nimbulus.xyz/dokuwiki/doku.php?id=gcc&rev=1745641387&do=diff Due to various hackery involving the kernel being built on gcc 12.3.0 but the packaged gcc in apt being 11.4.0, gcc was upgraded manually to 12.3.0. When using g++ however, it remains on 11.4. Make sure your build scripts account for this (alias cc=). text/html 2025-04-26T04:23:07+00:00 Anonymous (anonymous@undisclosed.example.com) hardware https://nimbulus.xyz/dokuwiki/doku.php?id=hardware&rev=1745641387&do=diff Hardware Description Compute nodes are 2x Xeon E5-2698 v3 (32 cores) with 128GB of ram each. 1 Nvidia P4 gpu is available per node (8GB of vram per node). Storage is provided locally (use your home directory) or on NFS mounts at /nfs and /scratchfs. text/html 2025-04-26T04:23:07+00:00 Anonymous (anonymous@undisclosed.example.com) nvidia https://nimbulus.xyz/dokuwiki/doku.php?id=nvidia&rev=1745641387&do=diff Introduction The nodes have an Nvidia Tesla P4 gpu installed for gpu-enabled computing. Cuda 12.4 is available by default. Using Python Using Nvidia is a real pain. There are two options: When submitting jobs through slurm, you can use computecanada's prebuilt modules via text/html 2025-09-04T15:57:48+00:00 Anonymous (anonymous@undisclosed.example.com) slurm https://nimbulus.xyz/dokuwiki/doku.php?id=slurm&rev=1757001468&do=diff In general submission should look like this: #!/bin/bash #SBATCH --cpus-per-task=(max 64) #SBATCH --mem=(max 124G) #SBATCH --time=(max 7-00:00) #SBATCH --output=%N-%j.out your commands here If you need a gpu, add this line: #SBATCH --gres=gpu:1 text/html 2025-04-26T04:23:07+00:00 Anonymous (anonymous@undisclosed.example.com) start https://nimbulus.xyz/dokuwiki/doku.php?id=start&rev=1745641387&do=diff Welcome to the documentation wiki of the Center for Low-Performance Computing! See important links * Hardware * cvmfs * slurm * nvidia * gcc