Sema, You could temporarily disable the enforcement of memory, in an effort to see what the daemons are reporting the job(s) as using.
Simply update the parameter "MemLimitEnforce" in your slurm.conf. I'd only recommend doing this for a very short period though. You definitely don't want compute nodes to start feeling memory pressure, leading to swapping. HTH, John DeSantis Sema Atasever wrote: > Hi Slurm-Dev, > > I have a *large dataset* stored as a text file. Consists of two separate > files (test and train) > > I am running into an error when training the SVM model. > > *Computing predictions by SVM...* > *slurmstepd: Job 3469 exceeded memory limit (4235584 > 2048000), being > killed* > *slurmstepd: Exceeded job memory limit* > > How can i fix this problem. > > Best. >
