Scaling gysela code beyond 32K-cores on bluegene/Q***

ESAIM: Proceedings and Surveys. 2013;43:117-135 DOI 10.1051/proc/201343008

 

Journal Homepage

Journal Title: ESAIM: Proceedings and Surveys

ISSN: 1270-900X (Print)

Publisher: EDP Sciences

Society/Institution: Société de Mathématiques Appliquées et Industrielles (SMAI)

LCC Subject Category: Technology: Technology (General): Industrial engineering. Management engineering: Applied mathematics. Quantitative methods | Science: Mathematics

Country of publisher: France

Language of fulltext: English

Full-text formats available: PDF

 

AUTHORS

Bigot J. (Maison de la simulation, CEA Saclay)
Grandgirard V. (CEA Cadarache)
Latu G. (CEA Cadarache)
Passeron Ch. (CEA Cadarache)
Rozar F.
Thomine O. (CEA Cadarache)

EDITORIAL INFORMATION

Peer review

Editorial Board

Instructions for authors

Time From Submission to Publication: 10 weeks

 

Abstract | Full Text

Gyrokinetic simulations lead to huge computational needs. Up to now, the semi- Lagrangian code Gysela performed large simulations using a few thousands cores (8k cores typically). Simulation with finer resolutions and with kinetic electrons are expected to increase those needs by a huge factor, providing a good example of applications requiring Exascale machines. This paper presents our work to improve Gysela in order to target an architecture that presents one possible way towards Exascale: the Blue Gene/Q. After analyzing the limitations of the code on this architecture, we have implemented three kinds of improvement: computational performance improvements, memory consumption improvements and disk i/o improvements. As a result, we show that the code now scales beyond 32k cores with much improved performances. This will make it possible to target the most powerful machines available and thus handle much larger physical cases.