With a $10 million dollar award from the National Science Foundation, San Diego Supercomputer Center (SDSC) at the University of California San Diego is procuring a new supercomputer, called Expanse, to be deployed next summer. Built by Dell EMC with Intel, Nvidia and Mellanox components, Expanse will double the capacity of SDSC’s existing NSF-funded resource, Comet, also built by Dell.
Like Comet, launched in 2015, Expanse will feature a heterogeneous CPU-GPU design and will focus on the needs of long-tail science. Citing its need for “productive computing,” SDSC says that Expanse will serve “virtually every discipline, from multi-messenger astronomy, genomics, and the social sciences [to…] earth sciences and biology.”
Expanse is projected to deliver a peak speed of 5 petaflops and to offer a 1.3 speedup for both CPU and GPU applications relative to Comet. According to the NSF award notice, Expanse will be organized into 13 SDSC Scalable Compute Units (SSCUs), each housing 56 CPU nodes and 4 GPU nodes along with over 60 TB of distributed non-volatile memory storage for user scratch. The SSCUs will be integrated with a 12-petabyte Lustre parallel file system and 7 petabytes of object storage. San Diego-based HPC storage provider Aeon is a vendor partner.
Mellanox is supplying HDR InfiniBand with a topology optimized for medium-scale parallelism (jobs of one to a few thousand cores).
The initial NSF award, valued at $10 million, runs from October 1, 2020 to September 30, 2025. A second smaller award directed at operations and user support will be made in the coming months.
The mission for expanse is two-fold. In addition to supporting traditional batch-oriented and science gateway computing, the system will also provide new capabilities to enable research increasingly dependent upon heterogeneous and distributed resources. “One of the key innovations in Expanse is its ability to support so-called composable systems at the continuum of computing with dynamic capabilities. Using tools such as Kubernetes, and workflow software we have developed over the years for projects including the NSF-funded WIFIRE and CHASE-CI programs, Expanse will extend the boundaries of what is possible by integration with the broader computational and data ecosystem,” said SDSC Chief Data Science Officer Ilkay Altintas, an Expanse co-PI and the director of SDSC’s Workflows for Data Science (WorDS) Center of Excellence.
“The name of our new system says it all,” said SDSC Director Michael Norman, the Principal Investigator (PI) for Expanse, and a computational astrophysicist. “As a standalone system, Expanse represents a substantial increase in the performance and throughput compared to our highly successful, NSF-funded Comet supercomputer. But with innovations in cloud integration and composable systems, as well as continued support for science gateways and distributed computing via the Open Science Grid, Expanse will allow researchers to push the boundaries of computing and answer questions previously not possible.”