When preparing to submit a job, keep the system's usable memory in mind and configure your job script to avoid oversubscribing the cores, which will degrade performance.
If your jobs approach the usable memory threshold shown as below, you may experience unexpected issues or job failures. Leave a margin of 2 or 3 percent.
System Total memory/node Usable memory/node Aaditya HPC 64 GB 60 GB
Also consider these suggestions:
- Spread memory-intensive jobs across more nodes than you would request for other jobs.
- Use fewer than the full number of cores on each node. This makes more memory available to run your processes.