NCSA’s Upcoming $10M Delta System to Expand Use of GPUs in Scientific Workloads

Print Friendly, PDF & Email

Delta, a new supercomputer to be deployed before the end of 2021 at the National Center for Supercomputing Application’s (NCSA), has as part of its mission the expanded adoption of GPU-accelerated scientific computing.

NCSA  Director Bill Gropp told us that that while NCSA is not new to GPUs (some years ago, staffers there configured a system using Sony PlayStation 2’s) the organization has decided to move forward with bringing accelerated computing into the mainstream of supercomputing workloads.

“I view it as a forward looking production system,” Gropp said of Delta, funded by a $10 million award from the National Science Foundation’s (NSF) HPC program. “In the end, it still has to support a broad spectrum of computational science applications, so there’s a limit to how radical it can be. That’s reflected in our configuration, which is heterogeneous in several directions. It does have CPUs and GPUs, but it also has a number of different kinds of nodes, which is intended to help applications make the transition from being just CPU codes to be able to use GPUs effectively.”

Bill Gropp of NCSA

While system components and anticipated performance numbers were not disclosed Gropp said Delta will incorporate Nvidia GPUs. NCSA, located at the University of Illinois, Urbana-Champaign, will integrate Delta into the national cyberinfrastructure ecosystem through the Extreme Science and Engineering Discovery Environment (XSEDE), an NSF-funded virtual organization that coordinates sharing of advanced digital services. Integration into XSEDE allows Delta to leverage the substantial portfolio of services and support offered therein and together deliver unprecedented advances in researcher productivity. This collaboration will promote synergy among multi-site workflows that include campus, national, and commercial cloud resources.

NCSA said the system will be a “balanced mixture” between next-generation CPU architectures and graphics processors, “making it the most performant GPU computing resource in NSF’s portfolio when it launches. Working with partners such as the Science Gateways Community Institute (SGCI) and accessibility researchers, it will change the way these advanced computing systems are used.”

The project is led by Principal Investigator (PI) Gropp and co-PIs and NCSA colleagues Gregory Bauer, Senior Technical Program Manager; Brett Bode, Assistant Director for the Blue Waters project; Timothy Boerner, Deputy Project Director for XSEDE; and Amy Schuele, Interim Associate Director of Integrated Cyberinfrastructure.

“With Delta, we will help emerging research areas, such as computational archaeology and digital agriculture, take advantage of new computing methods and hardware while simultaneously making advanced computing systems more usable and accessible to a broad community of researchers,” Gropp said.

“I view it as a forward looking production system. In the end, it still has to support a broad spectrum of computational science applications, so there’s a limit to how radical it can be. That’s reflected in our configuration, which is heterogeneous in several directions. It does have CPUs and GPUs, but it also has a number of different kinds of nodes, which is intended to help applications make the transition from being just CPU codes to be able to use GPUs effectively.” – Bill Gropp, director, NCSA

NCSA said Delta aligns with the organization’s commitment to diversity and seeks to promote a wider demographic among researchers by offering expanded training sessions and incorporating modern interfaces to improve usability. In addition, Delta will provide professional development opportunities to adapt research applications to more optimally use key features of the system.

“These awards represent a suite of complementary advanced computational capabilities and services aimed to empower new fundamental research in many fields,” says Amy Friedlander, Acting Director of NSF’s Office of Advanced Cyberinfrastructure. “NSF’s long-standing investments in advanced and innovative computing respond to the rapid evolution and expansion of computational- and data-intensive research being conducted across all of science and engineering.”