Reaching exascale: recent advances in computational fluid dynamics and their application to nuclear engineering by Prof. Elia Merzari (Penn State College of Engineering)

GPU-based supercomputing is enabling a significant advancement in Computational Fluid Dynamics (CFD) capabilities for nuclear reactors. In fact, GPU-based super-computers are allowing for the first time to perform full core CFD calculations with URANS and LES approaches. Key to this has been the development of NekRS, a novel GPU-oriented variant of Nek5000, an open source spectral element code in development at Argonne National Laboratory. NekRS delivers peak performance for key kernels on the GPUs, and good scaling performance even on GPU architectures. Recent simulations on the Frontier supercomputer demonstrate sustained performance on up to 72,000 GPUs. In this talk we review a few examples of this new and groundbreaking capabilities, including a the simulation of a full core pebble bed reactor. We also discuss how these calculations are being used to improve the fidelity of more traditional approaches such as porous media models for pebble beds. In fact, supercomputing simulation alone cannot impact design and safety analysis without a suitable multiscale framework.

Elia Merzari received his Ph.D. from Tokyo Institute of Technology with a thesis on the use of advanced computational fluid dynamics techniques to the simulation of flows in rod bundles. After graduation, he remained in the same institute as a Japanese Society for the Promotion of Science fellow. In 2009, he joined Argonne National Laboratory, first as a postdoctoral fellow and then as a regular staff in 2010. At Argonne, Dr. Merzari served in several roles in the Nuclear Energy Advanced modeling and simulation (NEAMS) program, for which he is currently the thermal-fluids lead. In 2019, he joined the faculty at Penn State as an associate professor. Dr. Merzari’s research relies on predictive large-scale simulations of turbulence to improve our physical understanding of complex flows and to ultimately design safer and more efficient nuclear reactors. He has received several awards related to these efforts in the area of high performance computing including: ANS Landis Young Member Engineering Achievement Award, ASME's George Westinghouse Silver Medal, and a R&D100 award.