The Data-Processing Shared Facility (DPSF) is now available to early-adopters. The DPSF is a new computational platform (developed from Hydra) which is complementary to the highly-successful Computational Shared Facility (CSF) which has been in production for several years: The DPSF is specified for high-memory and IO-intensive work.
The platform comprises:
- Compute nodes with either 256 GB or 512 GB of RAM;
- a /scratch filesystem of over 600 TB and, of course, has the same home-directories and research group shared areas as the CSF and other platforms.
Some limited free-at-the-point-of-use computation time is available.