Purdue University

A team of researchers has been using the Rosen Center for Advanced Computing (RCAC)’s powerful Anvil supercomputer to study gaseous jets in hypersonic crossflow, which is useful for understanding the operations of the systems that are used to steer aircraft and spacecraft in the upper atmosphere.

Joseph Jewell, an assistant professor of aeronautics and astronautics, collaborated with Deborah Levin, a professor of aerospace engineering at the University of Illinois at Urbana-Champaign, and her team of students, to compare their experimental results from Purdue’s hypersonic wind tunnel with computational simulations done on Anvil.

The simulation methods they use are computationally intensive but easily parallelizable, meaning they can make good use of Anvil’s many cores and massive computing power.

“This Anvil allocation has really helped us to do some work that we can put in proposals in the future,” says Jewell.

Anvil has helped the team make progress on answering questions about the degree to which jet interactions scale across different parameters, such as whether the penetration depth of the jet will change with pressure if the Mach number is consistent.

Access to Anvil is provided through the Advanced Cyberinfrastructure Coordination Ecosystem: Services & Support (ACCESS), previously the Extreme Science and Engineering Discovery Environment (XSEDE), in which Purdue has long been a partner. Purdue's participation in these national cyberinfrastructure programs has made it easier for faculty to benefit from their resources.

Anvil is Purdue University’s most powerful supercomputer yet, providing researchers from diverse backgrounds with advanced computing capabilities. Built through a $10 million system acquisition grant from the National Science Foundation (NSF), Anvil went into production in February of 2022 and will continue serving the research community through 2027.

Over the coming years, the Anvil supercomputer will continue to enable significant discoveries across many areas of science and engineering. Anvil will also serve as an experiential learning laboratory for students to gain real-world experience using computing for their science, and for student interns to work with the Anvil team to innovate and enhance Anvil’s capabilities.

Anvil, which was built in partnership with Dell and AMD, consists of 1,000 nodes with two 64-core AMD Epyc "Milan" processors, each with a peak performance of 5.3 petaflops. The supercomputer ecosystem also includes 32 large memory nodes, each with 1 TB of RAM, and 16 nodes each with four NVIDIA A100 Tensor Core GPUs providing 1.5 PF of single-precision performance to support machine learning and artificial intelligence applications.

Learn more about Anvil and its computational capacity here.