Investigating applications on the A64FX

Adrian Jackson; Michele Weiland; Nick Brown; Andrew Turner; Mark Parsons

doi:10.1109/CLUSTER49012.2020.00078

Investigating applications on the A64FX

Adrian Jackson, Michele Weiland, Nick Brown, Andrew Turner, Mark Parsons

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The A64FX processor from Fujitsu, being designed for computational simulation and machine learning applications, has the potential for unprecedented performance in HPC systems. In this paper, we evaluate the A64FX by benchmarking against a range of production HPC platforms that cover a number of processor technologies. We investigate the performance of complex scientific applications across multiple nodes, as well as single node and mini-kernel benchmarks. This paper finds that the performance of the A64FX processor across our chosen benchmarks often significantly exceeds other platforms, even without specific application optimisations for the processor instruction set or hardware. However, this is not true for all the benchmarks we have undertaken. Furthermore, the specific configuration of applications can have an impact on the runtime and performance experienced.

Original language	English
Title of host publication	2020 IEEE International Conference on Cluster Computing (CLUSTER)
Publisher	Institute of Electrical and Electronics Engineers
Pages	549-558
Number of pages	10
ISBN (Electronic)	978-1-7281-6677-3
ISBN (Print)	978-1-7281-6678-0
DOIs	https://doi.org/10.1109/CLUSTER49012.2020.00078
Publication status	Published - 2 Nov 2020
Event	IEEE Cluster 2020 - Kobe, Japan Duration: 14 Sept 2020 → 17 Sept 2020 https://clustercomp.org/2020/

Publication series

Name
Publisher	IEEE
ISSN (Print)	1552-5244
ISSN (Electronic)	2168-9253

Conference

Conference	IEEE Cluster 2020
Abbreviated title	Cluster2020
Country/Territory	Japan
City	Kobe
Period	14/09/20 → 17/09/20
Internet address	https://clustercomp.org/2020/

Access to Document

10.1109/CLUSTER49012.2020.00078Licence: All Rights Reserved

Investigating Applications_JACKSON_DOA14082020_AFVAccepted author manuscript, 487 KBLicence: Creative Commons: Attribution-NonCommercial-ShareAlike (CC BY-NC-SA)

https://ieeexplore.ieee.org/document/9229611Licence: All Rights Reserved

EPCC Tier 2 HPC Service
Parsons, M.
EPSRC
1/11/16 → 30/04/21
Project: Research

Cite this

@inproceedings{d543333ca874487f9d255427989e1c78,

title = "Investigating applications on the A64FX",

abstract = "The A64FX processor from Fujitsu, being designed for computational simulation and machine learning applications, has the potential for unprecedented performance in HPC systems. In this paper, we evaluate the A64FX by benchmarking against a range of production HPC platforms that cover a number of processor technologies. We investigate the performance of complex scientific applications across multiple nodes, as well as single node and mini-kernel benchmarks. This paper finds that the performance of the A64FX processor across our chosen benchmarks often significantly exceeds other platforms, even without specific application optimisations for the processor instruction set or hardware. However, this is not true for all the benchmarks we have undertaken. Furthermore, the specific configuration of applications can have an impact on the runtime and performance experienced.",

author = "Adrian Jackson and Michele Weiland and Nick Brown and Andrew Turner and Mark Parsons",

year = "2020",

month = nov,

day = "2",

doi = "10.1109/CLUSTER49012.2020.00078",

language = "English",

isbn = "978-1-7281-6678-0",

publisher = "Institute of Electrical and Electronics Engineers",

pages = "549--558",

booktitle = "2020 IEEE International Conference on Cluster Computing (CLUSTER)",

address = "United States",

note = "IEEE Cluster 2020, Cluster2020 ; Conference date: 14-09-2020 Through 17-09-2020",

url = "https://clustercomp.org/2020/",

}

TY - GEN

T1 - Investigating applications on the A64FX

AU - Jackson, Adrian

AU - Weiland, Michele

AU - Brown, Nick

AU - Turner, Andrew

AU - Parsons, Mark

PY - 2020/11/2

Y1 - 2020/11/2

N2 - The A64FX processor from Fujitsu, being designed for computational simulation and machine learning applications, has the potential for unprecedented performance in HPC systems. In this paper, we evaluate the A64FX by benchmarking against a range of production HPC platforms that cover a number of processor technologies. We investigate the performance of complex scientific applications across multiple nodes, as well as single node and mini-kernel benchmarks. This paper finds that the performance of the A64FX processor across our chosen benchmarks often significantly exceeds other platforms, even without specific application optimisations for the processor instruction set or hardware. However, this is not true for all the benchmarks we have undertaken. Furthermore, the specific configuration of applications can have an impact on the runtime and performance experienced.

AB - The A64FX processor from Fujitsu, being designed for computational simulation and machine learning applications, has the potential for unprecedented performance in HPC systems. In this paper, we evaluate the A64FX by benchmarking against a range of production HPC platforms that cover a number of processor technologies. We investigate the performance of complex scientific applications across multiple nodes, as well as single node and mini-kernel benchmarks. This paper finds that the performance of the A64FX processor across our chosen benchmarks often significantly exceeds other platforms, even without specific application optimisations for the processor instruction set or hardware. However, this is not true for all the benchmarks we have undertaken. Furthermore, the specific configuration of applications can have an impact on the runtime and performance experienced.

U2 - 10.1109/CLUSTER49012.2020.00078

DO - 10.1109/CLUSTER49012.2020.00078

M3 - Conference contribution

SN - 978-1-7281-6678-0

SP - 549

EP - 558

BT - 2020 IEEE International Conference on Cluster Computing (CLUSTER)

PB - Institute of Electrical and Electronics Engineers

T2 - IEEE Cluster 2020

Y2 - 14 September 2020 through 17 September 2020

ER -

Investigating applications on the A64FX

Abstract

Publication series

Conference

Access to Document

Fingerprint

Projects

EPCC Tier 2 HPC Service

Cite this