Science Success Story
PSC's Bridges-2 and SDSC's Expanse Now Allocable by XSEDE
XSEDE Welcomes Two New Level 1 Service Providers
Two relatively new NSF-funded supercomputers, the Bridges-2 platform at the Pittsburgh Supercomputing Center (PSC), and Expanse at the San Diego Supercomputer Center (SDSC), are available for allocation requests via XSEDE. More information about each individual resource may be found below. Specific inquiries may be directed to firstname.lastname@example.org.
Bridges-2 is an NSF Category I (Award #1928147) supercomputing platform designed to provide researchers with massive computational capacity and the flexibility to adapt to the rapidly evolving field of data- and computation-intensive research, particularly in the confluence of Big Data and artificial intelligence. Bridges-2 will create opportunities for collaboration and convergence research. It supports both traditional and rapidly evolving research communities and applications. Bridges-2 will continue to integrate new technologies for converged, scalable HPC, artificial intelligence and data. The system will also prioritize researcher productivity and ease of use, providing an extensible architecture for interoperation with complementary data-intensive projects, campus resources, and clouds. Bridges-2 comes as a replacement for PSC's former Bridges machine, which was officially retired on February 15, 2021. Users who wish to gain access to Bridges-2 should make a request in the upcoming March 15-April 15 XSEDE Research Allocation request period.
Expanse, also an NSF Category I Award (#1928224), is a new supercomputer designed to advance research that is increasingly dependent upon heterogeneous and distributed resources. Expanse entered production on December 7, 2020. With a peak speed of 5 Petaflop/s, Expanse nearly doubles the performance of its SDSC forerunner, Comet (slated for retirement in July 2021), with next-generation processors and NVIDIA's GPUs. Expanse increases throughput of real-world workloads by a factor of at least 1.3 for both CPU and GPU applications relative to Comet, while supporting an even larger and more diverse research community. Expanse's GPU-accelerated nodes provide a much-needed GPU capability to the user community, serving both well-established applications in areas such as molecular dynamics as well as rapidly growing demand for resources to support machine learning and artificial intelligence. Expanse also offers two innovative capabilities: support for direct integration with public cloud; and composable systems. Users interested in these should contact XSEDE for additional information. Users who wish to try Expanse can request allocations through normal XSEDE allocation channels, including the upcoming March 15 - April 15 allocation cycle. Users transitioning from Comet should consult the Expanse documentation as the system is the first XSEDE system to offer the new AMD EPYC 7742 (Rome) processors. More information is also available in the Comet to Expanse Transition Tutorial, and in the upcoming Comet to Expanse Transition Tutorial on March 4, 2021.