Poster Type: Research Posters
Author: Sambit Mishra (Texas A&M University), Semih Akkurt (NVIDIA Corporation), Freddie Witherden (Texas A&M University)
Abstract: Scale-resolving simulations performed on clusters can be expensive, spanning hours to months. Efficient use of shared HPC resources increasingly involves allocations combining devices with different throughputs, requiring effective load partitioning across heterogeneous architectures. This work develops an online, distributed, diffusion-based repartitioning strategy driven by observed MPI wait-time statistics at runtime. The method infers cost imbalance to derive a target load distribution, then incrementally relocates interface-adjacent elements across ranks while preserving the existing partition structure. Integrated into the open-source compressible Navier–Stokes solver PyFR, the method is evaluated for flow around a circular cylinder at Re = 3900. Online load balancing achieves 101%, 116%, and 119% of homogeneous sum-of-parts performance targets across three heterogeneous configurations. Completed RK45 simulations achieve 2.4x, 2.8x, and 7.7x time-to-solution speedups.
Best Poster Finalist (BP): no
Poster: PDF
Poster Summary: PDF