NSF Award
2322260
This project to the University of Colorado and partners in the Rocky Mountain Advanced Computing Consortium (RMACC) expands the capacity of the Alpine Computing Cluster, a regional resource for RMACC. The project expands capacity by more than 15% and triples the number of high-memory nodes. A wide range of research is supported, including physics, astronomy, engineering, biology, and atmospheric science. The project supports diverse workloads for the approximately 30 member institutions of RMACC. Student engagement is strong, as ten students from across the RMACC region are part of a cohort working closely with the system administration team to learn how to navigate the design, procurement, deployment, and administration of the system. The project supports researchers in EPSCoR jurisdictions, at Primarily Undergraduate Institutions, Hispanic Serving Institutions, and institutions primarily serving rural communities within RMACC.<br/><br/>As of February 2023, the Alpine cluster comprises 18,080 AMD Milan CPU cores as well as 60 GPU accelerators (36xNVIDIA A100’s, 24xAMD MI100’s). The 317 nodes are interconnected through 25Gb Ethernet, with standard compute nodes additionally featuring HDR100 InfiniBand links. Alpine is also equipped with a 1.9PB GPFS scratch filesystem. The expansion includes 32 nodes of 64C AMD Milan CPUs and 256GB of memory as well as eight nodes with 64C AMD Milan CPUs and 1TB of memory. It also includes two InfiniBand switches and associated HDR cables for high-throughput, low-latency networking. The expansion will provide an additional 1.3PB to Alpine’s high-speed scratch filesystem to meet the growing demand for data intensive computing research.<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
