Using Julia on Cluster

[legendaryh]

I am new to HPC. I am trying to run Julia on a cluster that uses slurm, but there’s no julia module pre-installed in it. Although I have installed Julia in my directory, I don’t know how to write the sbatch file that calls julia from my directory and then runs the .jl file. Anyone have any idea how to do it?

[KajWiik]

See e.g.

Running Julia in a SLURM Cluster

I’m looking to provide an explicit explanation of the simplest way to get a Julia program running in parallel across multiple nodes in a SLURM cluster (basically a minimal working example that illustrates the logic). My impression so far is that there are two primary ways to run Julia in a SLURM cluster. Suppose I want to define a function and run it in a parallel for-loop on N cores distributed across M nodes in the cluster: Option 1 The first option comes from this Stackoverflow post. Basic…

[carstenbauer]

What’s unclear to you? Something like the following should do it.

#!/bin/bash
#SBATCH -t 1:00:00
#SBATCH -N 1
#SBATCH -n 1
#SBATCH -J "my julia computation"
#SBATCH -p normal
#SBATCH -A myaccount

export PATH=/path/to/my/custom/julia/:$PATH
srun julia myscript.jl

Of course, you likely need not change the sbatch parameters, like the partition and account name.