OpenACC acceleration for the $P_N\text{–}{P}_{N\text{-}2}$ algorithm in Nek5000

Journal Article · 28 May 2019 · Journal of Parallel and Distributed Computing

DOI: https://doi.org/10.1016/j.jpdc.2019.05.010 · OSTI ID:1571263

Otero, Evelyn ^[1]; Gong, Jing ^[1]; Min, Misun ^[2]; Fischer, Paul ^[2]; Schlatter, Philipp ^[1]; Laure, Erwin ^[1]

---

1. KTH Royal Inst. of Technology, Stockholm (Sweden)
2. Argonne National Lab. (ANL), Argonne, IL (United States)

Due to its high performance and throughput capabilities, GPU-accelerated computing is becoming a popular technology in scientific computing, in particular using programming models such as CUDA and OpenACC. The main advantage with OpenACC is that it enables to simply port codes in their ‘‘original’’ form to GPU systems through compiler directives, thus allowing an incremental approach. An OpenACC implementation is applied to the CFD code Nek5000 for simulation of incompressible flows, based on the spectral-element method. The work follows up previous implementations and focuses now on the P_N- P_N-2 method for the spatial discretization of the Navier–Stokes equations. Performance results of the ported code show a speed-up of up to 3.1 on multi-GPU for a polynomial order N > 11.

• Research Organization: Argonne National Laboratory (ANL), Argonne, IL (United States)
• Sponsoring Organization: Swedish Foundation for Strategic Research (SSF); USDOE
• Grant/Contract Number: AC02-06CH11357
• OSTI ID: 1571263
• Journal Information: Journal of Parallel and Distributed Computing, Journal Name: Journal of Parallel and Distributed Computing Journal Issue: C Vol. 132; ISSN 0743-7315
• Publisher: ElsevierCopyright Statement
• Country of Publication: United States
• Language: English