RUS  ENG
Full version
JOURNALS // Preprints of the Keldysh Institute of Applied Mathematics // Archive

Keldysh Institute preprints, 2018 152, 20 pp. (Mi ipmp2511)

This article is cited in 1 paper

Performance of Elbrus-8C CPU in supercomputer CFD applications

A. V. Gorobetsa, M. I. Neiman-Zadeb, S. K. Okunevb, A. A. Kalyakinb, S. A. Sukova

a Keldysh Institute of Applied Mathematics of Russian Academy of Sciences, Moscow
b АО "МЦСТ", 119991, Москва, Ленинский проспект, д. 51

Abstract: The present work is devoted to performance evaluation of a multicore CPU Elbrus-8C in supercomputer computational fluid dynamics (CFD) applications. Parallel simulation codes based on higher-accuracy methods on unstructured meshes for modelling of turbulent flows are considered. Main features of the Elbrus architecture are described. Approaches for adaptation and optimization of computing software are presented. Performance has been investigated for the whole algorithms and for their main operations separately. Benchmarking results in comparison with various Intel and AMD multicore CPUs are presented.

Keywords: HPC, Elbrus processor, CFD, supercomputer simulations.

DOI: 10.20948/prepr-2018-152



Bibliographic databases:


© Steklov Math. Inst. of RAS, 2024