NYU High Performance Computing

NYU High Performance Computing (NYU HPC) provides access to state-of-the-art supercomputer hardware and cloud services to eligible faculty and students across all of NYU. The flagship system is the Greene cluster, named after Greene Street in SoHo, which is one of the most powerful supercomputers in higher education and among the top 100 greenest supercomputers in the world.

Comparison: SCRC vs NYU HPC

For SCRC users considering NYU HPC, here are key differences:

Feature        | SCRC                      | NYU HPC Greene
---------------|---------------------------|------------------------------------
Scale          | Moderate-sized cluster    | Large-scale supercomputer
Access         | Stern community focus     | University-wide access
GPUs           | Limited GPU resources     | 332 high-end GPUs (RTX8000, V100)
Storage        | Local storage solutions   | 7.3PB enterprise storage
Queue Times    | Generally shorter         | Can be longer for popular resources
Support        | Personalized SCRC support | University-wide HPC support
Learning Curve | Moderate                  | Steeper (more complex environment)

When to Use NYU HPC

• Large-scale computations requiring significant resources

• GPU-intensive workloads needing multiple high-end GPUs

• Big data processing with massive datasets

• Long-running jobs that exceed SCRC capacity

When to Use SCRC

• Getting started with HPC computing

• Moderate computational needs within SCRC capacity

• Quick turnaround on smaller jobs

• Direct support from SCRC team

• Stern-specific workflows and requirements

Both systems complement each other, and researchers may use both depending on their computational needs and project requirements.