[en] ARM architecture central processing units are increasingly prevalent in high performance computers due to their energy efficiency, scalability and cost-effectiveness. The overall goal of this study is to evaluate the suitability of ARM-based cloud computing instances for executing finite element computations. Specifically, we show performance results executing the FEniCS Project finite element software on Amazon Web Services (AWS) c7g and c7gn instances with Graviton3 processors. These processors support ARMv8.4-A instruction set with Scalable Vector Extensions (SVE) for Single Instruction Multiple Data operations and the Elastic Fabric Adaptor for communications between instances. Both clang 18 and GCC 13 compilers successfully generated optimized code using SVE instructions which ensures that users can achieve optimized performance without extensive manual tuning. Testing a distributed memory parallel DOLFINx Poisson solver with up to 512 Message Passing Interface processes, we found that the performance and scalability of the AWS instances are comparable to a dedicated AMD EPYC Rome cluster installed at the University of Luxembourg. These findings demonstrate that ARM-based cloud computing instances, exemplified by AWS Graviton3, can be competitive for distributed memory parallel finite element analysis.
Research center :
ULHPC - University of Luxembourg: High Performance Computing
Disciplines :
Computer science Engineering, computing & technology: Multidisciplinary, general & others
Author, co-author :
HABERA, Michal ; University of Luxembourg > Faculty of Science, Technology and Medicine > Department of Engineering > Team Andreas ZILIAN
HALE, Jack ; University of Luxembourg > Faculty of Science, Technology and Medicine (FSTM) > Department of Engineering (DoE)
External co-authors :
no
Language :
English
Title :
The FEniCS Project on AWS Graviton3
Publication date :
In press
Event name :
FEniCS 2024
Event organizer :
Simula Research Laboratory
Event place :
Oslo, Norway
Event date :
12 to 14 June 2024
Audience :
International
Main work title :
FEniCS 2024 Conference Proceedings
Publisher :
Springer-Verlag
Collection name :
Simula SpringerBriefs on Computing
Peer reviewed :
Peer reviewed
Focus Area :
Computational Sciences
FnR Project :
FNR17205623 - Constraint Aware Optimization Of Topology In Design-for-additive-manufacturing, 2022 (01/11/2022-31/10/2024) - Michal Habera
Funders :
FNR - Fonds National de la Recherche
Funding text :
This research was funded in whole, or in part, by the National Research Fund (FNR), grant reference COAT/17205623. For the purpose of open access, and in fulfillment of the obligations arising from the grant agreement, the author has applied a Creative Commons Attribution 4.0 International (CC BY 4.0) license to any Author Accepted Manuscript version arising from this submission.
