With 24 TiB of shared memory, the SDF is well suited to an important class of problems that are difficult or impossible to run on conventional distributed-memory clusters. Use cases include analyses on very large graphs that must remain resident in memory; large-scale bioinformatics pipelines; and irregular or sparse computations. It is also a strong fit for in-memory AI-readiness preparation of very large datasets for downstream use on the Neocortex CS-3 or on Bridges-2.
We are looking to support research efforts where this capability can make a meaningful difference. Successful applicants will be paired with a Neocortex collaborator who will help them set up their working environment and run their project on the system.
Applications will be accepted via Easychair on a rolling basis, from May 18 to June 7, 2026.
For more information and to submit your proposal, please visit Call for Proposals: NeoSDF 2026 | PSC .
To ask questions and discuss the details contact the Neocortex team at neocortex [at] psc.edu.