The current infrastructure comprises:
24 compute nodes, each with:
- 32 physical cores
- 512 GB RAM
- 1 TB working-job (volatile-scratch) space
- 200 TB NFS no-backup scratch NFS storage for larger data set
Job Management is handled by Univa Grid Engine.
A recently approved expansion will see posthaste the integration of an additional eight compute nodes, as well as 12 nVIDIA Tesla P6 GPUs.
The "cluster" mounts multi-TB, NFS-provided storage, which serves both persistent-scratch data and persistent-store data (not backed up; backed up, respectively).