Performance-portable Acceleration of a DNS Turbulent Reacting Flows Application with OPS

Read full text Summarise with VIRAT
Authors: Ashutosh Londhe, Hamid Kavari, Vishnu Mohan, J Kaushal, Istvan Zoltan Reguly, Nilanjan Chakraborty, R. Stewart Cant, Gihan R. Mudalige
Cited by: 0
References: 18

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

SENGA+ is a high-order finite difference compressible direct numerical simulation (DNS) code for simulating turbulent reacting flows. It incorporates detailed chemical reactions and transport with high order numerical schemes to achieve high-fidelity simulations and excellent parallel scaling. This paper presents efforts to deploy SENGA+ on next-generation high-performance computing systems, using the OPS DSL. Key challenges in the process are presented, underpinned by specialized optimizations to gain near-optimal performance on multicore/ many-core systems. OPS enables automatic generation of multiple parallelizations, highly customized for the target architectures, reducing developer effort and increasing code longevity. Performance evaluation demonstrates up to 22× reduction in runtime on a GPU node with 4 AMD MI250X compared to the original CPU-only code. We present validation simulations using the re-engineered application with near-optimal scaling on up to 2k GPUs. The new code enables simulating production level combustion problems at high-fidelity within tractable timeframes that were previously prohibitively expensive.

Fields of Study & Subjects

Parallel Computing and Optimization Techniques Embedded Systems Design Techniques Computer Graphics and Visualization Techniques

Publication details

Journal / source
IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum
Pages
310–320
Publication date
1 May 2026
Language
EN