Heterogeneous Programming for the Homogeneous Majority

Tom Deakin; Jim H Cownie; Tom Lin; Simon N McIntosh-Smith

doi:10.1109/P3HPC56579.2022.00006

Heterogeneous Programming for the Homogeneous Majority

Tom Deakin, Jim H Cownie, Tom Lin, Simon N McIntosh-Smith

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

6 Citations (Scopus)

263 Downloads (Pure)

Abstract

In order to take advantage of the burgeoning diversity in processors at the frontier of supercomputing, the HPC community is migrating and improving codes to utilise heterogeneous nodes, where accelerators, principally GPUs, are highly prevalent in top-tier supercomputer designs. Programs therefore need to embrace at least some of the complexities of heterogeneous architectures. Parallel programming models have evolved to express heterogeneous paradigms whilst providing mechanisms for writing portable, performant programs. History shows that technologies first introduced at the frontier percolate down to local workhorse systems. However, we expect there will always be a mix of systems, some heterogeneous, but some remaining as homogeneous CPU systems. Thus it is important to ensure codes adapted for heterogeneous systems continue to run efficiently on CPUs. In this study, we explore how well widely used heterogeneous programming models perform on CPU-only platforms, and survey the performance portability they offer on the latest CPU architectures.

Original language	English
Title of host publication	2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	1-13
Number of pages	13
ISBN (Electronic)	978-1-6654-6021-7
ISBN (Print)	978-1-6654-6022-4
DOIs	https://doi.org/10.1109/P3HPC56579.2022.00006
Publication status	E-pub ahead of print - 30 Jan 2023
Event	2022 International Workshop on Performance Portability and Productivity (P3HPC) - Duration: 13 Nov 2022 → 13 Nov 2022

Publication series

Name
Publisher	IEEE
ISSN (Print)	2831-3895
ISSN (Electronic)	2831-3909

Conference

Conference	2022 International Workshop on Performance Portability and Productivity (P3HPC)
Period	13/11/22 → 13/11/22

Bibliographical note

Funding Information:
For the purpose of open access, the author(s) has applied a Creative Commons Attribution (CC BY) license to any Accepted Manuscript version arising. This work was in part funded by EPSRC through the Strategic Partnership in Computational Science for Advanced Simulation and Modelling of Engineering Systems (ASiMoV) project, EP/S005072/1.

Funding Information:
This work used the Isambard 2 UK National Tier-2 HPC Service (http://gw4.ac.uk/isambard/) operated by GW4 and the UK Met Office, and funded by EPSRC (EP/T022078/1). The University of Bristol is an Intel oneAPI Center of Excellence, which helped support this work. Thanks to AWS for supporting access to Graviton 3.

Publisher Copyright:
© 2022 IEEE.

Access to Document

10.1109/P3HPC56579.2022.00006

Full-text PDF (accepted author manuscript)
This is the accepted author manuscript (AAM). The final published version (version of record) is available online via IEEE at https://ieeexplore.ieee.org/document/10024620. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 376 KBLicence: CC BY

Handle.net

Persistent link

Cite this

@inproceedings{b333d9419fea44c583bb541956881e84,

title = "Heterogeneous Programming for the Homogeneous Majority",

abstract = "In order to take advantage of the burgeoning diversity in processors at the frontier of supercomputing, the HPC community is migrating and improving codes to utilise heterogeneous nodes, where accelerators, principally GPUs, are highly prevalent in top-tier supercomputer designs. Programs therefore need to embrace at least some of the complexities of heterogeneous architectures. Parallel programming models have evolved to express heterogeneous paradigms whilst providing mechanisms for writing portable, performant programs. History shows that technologies first introduced at the frontier percolate down to local workhorse systems. However, we expect there will always be a mix of systems, some heterogeneous, but some remaining as homogeneous CPU systems. Thus it is important to ensure codes adapted for heterogeneous systems continue to run efficiently on CPUs. In this study, we explore how well widely used heterogeneous programming models perform on CPU-only platforms, and survey the performance portability they offer on the latest CPU architectures.",

author = "Tom Deakin and Cownie, {Jim H} and Tom Lin and McIntosh-Smith, {Simon N}",

note = "Funding Information: For the purpose of open access, the author(s) has applied a Creative Commons Attribution (CC BY) license to any Accepted Manuscript version arising. This work was in part funded by EPSRC through the Strategic Partnership in Computational Science for Advanced Simulation and Modelling of Engineering Systems (ASiMoV) project, EP/S005072/1. Funding Information: This work used the Isambard 2 UK National Tier-2 HPC Service (http://gw4.ac.uk/isambard/) operated by GW4 and the UK Met Office, and funded by EPSRC (EP/T022078/1). The University of Bristol is an Intel oneAPI Center of Excellence, which helped support this work. Thanks to AWS for supporting access to Graviton 3. Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 International Workshop on Performance Portability and Productivity (P3HPC) ; Conference date: 13-11-2022 Through 13-11-2022",

year = "2023",

month = jan,

day = "30",

doi = "10.1109/P3HPC56579.2022.00006",

language = "English",

isbn = "978-1-6654-6022-4",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

pages = "1--13",

booktitle = "2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)",

address = "United States",

}

Deakin, T, Cownie, JH, Lin, T & McIntosh-Smith, SN 2023, Heterogeneous Programming for the Homogeneous Majority. in 2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC). Institute of Electrical and Electronics Engineers (IEEE), pp. 1-13, 2022 International Workshop on Performance Portability and Productivity (P3HPC), 13/11/22. https://doi.org/10.1109/P3HPC56579.2022.00006

Heterogeneous Programming for the Homogeneous Majority. / Deakin, Tom; Cownie, Jim H; Lin, Tom et al.
2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC). Institute of Electrical and Electronics Engineers (IEEE), 2023. p. 1-13.

Research output: Chapter in Book/Report/Conference proceeding › Conference Contribution (Conference Proceeding)

TY - GEN

T1 - Heterogeneous Programming for the Homogeneous Majority

AU - Deakin, Tom

AU - Cownie, Jim H

AU - Lin, Tom

AU - McIntosh-Smith, Simon N

N1 - Funding Information: For the purpose of open access, the author(s) has applied a Creative Commons Attribution (CC BY) license to any Accepted Manuscript version arising. This work was in part funded by EPSRC through the Strategic Partnership in Computational Science for Advanced Simulation and Modelling of Engineering Systems (ASiMoV) project, EP/S005072/1. Funding Information: This work used the Isambard 2 UK National Tier-2 HPC Service (http://gw4.ac.uk/isambard/) operated by GW4 and the UK Met Office, and funded by EPSRC (EP/T022078/1). The University of Bristol is an Intel oneAPI Center of Excellence, which helped support this work. Thanks to AWS for supporting access to Graviton 3. Publisher Copyright: © 2022 IEEE.

PY - 2023/1/30

Y1 - 2023/1/30

N2 - In order to take advantage of the burgeoning diversity in processors at the frontier of supercomputing, the HPC community is migrating and improving codes to utilise heterogeneous nodes, where accelerators, principally GPUs, are highly prevalent in top-tier supercomputer designs. Programs therefore need to embrace at least some of the complexities of heterogeneous architectures. Parallel programming models have evolved to express heterogeneous paradigms whilst providing mechanisms for writing portable, performant programs. History shows that technologies first introduced at the frontier percolate down to local workhorse systems. However, we expect there will always be a mix of systems, some heterogeneous, but some remaining as homogeneous CPU systems. Thus it is important to ensure codes adapted for heterogeneous systems continue to run efficiently on CPUs. In this study, we explore how well widely used heterogeneous programming models perform on CPU-only platforms, and survey the performance portability they offer on the latest CPU architectures.

AB - In order to take advantage of the burgeoning diversity in processors at the frontier of supercomputing, the HPC community is migrating and improving codes to utilise heterogeneous nodes, where accelerators, principally GPUs, are highly prevalent in top-tier supercomputer designs. Programs therefore need to embrace at least some of the complexities of heterogeneous architectures. Parallel programming models have evolved to express heterogeneous paradigms whilst providing mechanisms for writing portable, performant programs. History shows that technologies first introduced at the frontier percolate down to local workhorse systems. However, we expect there will always be a mix of systems, some heterogeneous, but some remaining as homogeneous CPU systems. Thus it is important to ensure codes adapted for heterogeneous systems continue to run efficiently on CPUs. In this study, we explore how well widely used heterogeneous programming models perform on CPU-only platforms, and survey the performance portability they offer on the latest CPU architectures.

U2 - 10.1109/P3HPC56579.2022.00006

DO - 10.1109/P3HPC56579.2022.00006

M3 - Conference Contribution (Conference Proceeding)

SN - 978-1-6654-6022-4

SP - 1

EP - 13

BT - 2022 IEEE/ACM International Workshop on Performance, Portability and Productivity in HPC (P3HPC)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 2022 International Workshop on Performance Portability and Productivity (P3HPC)

Y2 - 13 November 2022 through 13 November 2022

ER -

Heterogeneous Programming for the Homogeneous Majority

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Handle.net

Fingerprint

Cite this