Spent last week fighting with issues on a slurm cluster, and having finally figured it out, I wanted to share the result (and a warning): I had been parallelizing through a slurm batch script with this call: julia --machinefile $SLURM_NODEFILE indiv_array.jl (For full script, see here ) Turns out…

Using srun should be fine since that is the correct way of starting a job. The workflow should work something like this: salloc | sbatch # create resources. julia> addprocs(SlurmManager(2)) # SlurmManager should inherit the outside allocation.

Issues with machinefile and SLURM

Specific Domains Julia at Scale

nickeubank December 20, 2017, 7:41pm 3

OH! So an srun executed inside a slurm allocation doesn’t try to create a new allocaiton; it start processes in that existing allocation?

Topic		Replies	Views
Configure Slurm File for Parallelization Julia at Scale parallel , cluster , distributed , slurm	4	95	October 13, 2024
SLURM manager: one node with multiple tasks General Usage slurm	2	121	December 28, 2024
Debugging possible issue with `machinefile` option on SLURM system Julia at Scale	9	1693	December 19, 2017
Julia on Slurm Cluster Specific Domains question	5	556	March 13, 2021
SLURM: Julia scripts do not exit or produce any output New to Julia cluster	9	1857	September 29, 2022

Issues with machinefile and SLURM

Related topics