Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems

Tom Lin; Simon N McIntosh-Smith; Tom Deakin

doi:10.1109/PMBS56514.2022.00009

Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems

Tom Lin, Simon N McIntosh-Smith, Tom Deakin

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

16 Citations (Scopus)

373 Downloads (Pure)

Abstract

Recent revisions to the ISO C++ standard have added specifications for parallel algorithms. These additions cover common use-cases, including sequence traversal, reduction, and even sorting, many of which are highly applicable in HPC, and thus represent a potential for increased performance and productivity.

This study evaluates the state of the art for implementing heterogeneous HPC applications using the latest built-in ISO C++17 parallel algorithms. We implement C++17 ports of representative HPC mini-apps that cover both compute-bound and memory bandwidth-bound applications. We then conduct benchmarks on CPUs and GPUs, comparing our ports to other widely-available parallel programming models, such as OpenMP, CUDA, and SYCL.

Finally, we show that C++17 parallel algorithms are able to achieve competitive performance across multiple mini-apps on many platforms, with some notable exceptions. We also discuss several key topics, including portability, and describe workarounds for a number of remaining issues, including index-based traversal and accelerator device/memory management.

Original language	English
Title of host publication	2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)
Publisher	Institute of Electrical and Electronics Engineers (IEEE)
Pages	36-47
Number of pages	12
ISBN (Electronic)	9781665451857
ISBN (Print)	9781665451864
DOIs	https://doi.org/10.1109/PMBS56514.2022.00009
Publication status	Published - 30 Jan 2023
Event	2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS) - Dallas, United States Duration: 14 Dec 2022 → … https://www.dcs.warwick.ac.uk/pmbs/pmbs/PMBS/Welcome.html

Conference

Conference	2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)
Abbreviated title	PMBS22
Country/Territory	United States
City	Dallas
Period	14/12/22 → …
Internet address	https://www.dcs.warwick.ac.uk/pmbs/pmbs/PMBS/Welcome.html

Bibliographical note

Funding Information:
ACKNOWLEDGMENT This work makes use of the following services: Isambard UK National Tier-2 HPC Service (https://gw4.ac.uk/isambard) operated by GW4 and the UK Met Office, funded by EPSRC (EP/P020224/1); HPC Zoo, a research cluster managed by the HPC Group at the University of Bristol (https://uob-hpc.github.io/zoo); Intel DevCloud, an online cluster for developers (https://devcloud.intel.com);EC2 Graviton instances, with access supported by AWS.

Publisher Copyright:
© 2022 IEEE.

Keywords

Performance Portability
Programming Models
GPUs
C++17
PSTL
stdpar

Access to Document

10.1109/PMBS56514.2022.00009

Full-text PDF (accepted manuscript)
This is the accepted author manuscript (AAM). The final published version (version of record) is available online via IEEE at https://doi.org/10.1109/PMBS56514.2022.00009. Please refer to any applicable terms of use of the publisher.
Accepted author manuscript, 382 KB

Handle.net

Persistent link

Cite this

Lin, T., McIntosh-Smith, S. N., & Deakin, T. (2023). Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems. In 2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS) (pp. 36-47). Institute of Electrical and Electronics Engineers (IEEE). https://doi.org/10.1109/PMBS56514.2022.00009

@inproceedings{d0c029ffa09f4aa099b43a43728f5bc4,

title = "Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems",

abstract = "Recent revisions to the ISO C++ standard have added specifications for parallel algorithms. These additions cover common use-cases, including sequence traversal, reduction, and even sorting, many of which are highly applicable in HPC, and thus represent a potential for increased performance and productivity. This study evaluates the state of the art for implementing heterogeneous HPC applications using the latest built-in ISO C++17 parallel algorithms. We implement C++17 ports of representative HPC mini-apps that cover both compute-bound and memory bandwidth-bound applications. We then conduct benchmarks on CPUs and GPUs, comparing our ports to other widely-available parallel programming models, such as OpenMP, CUDA, and SYCL. Finally, we show that C++17 parallel algorithms are able to achieve competitive performance across multiple mini-apps on many platforms, with some notable exceptions. We also discuss several key topics, including portability, and describe workarounds for a number of remaining issues, including index-based traversal and accelerator device/memory management. ",

keywords = "Performance Portability, Programming Models, GPUs, C++17, PSTL, stdpar",

author = "Tom Lin and McIntosh-Smith, \{Simon N\} and Tom Deakin",

note = "Funding Information: ACKNOWLEDGMENT This work makes use of the following services: Isambard UK National Tier-2 HPC Service (https://gw4.ac.uk/isambard) operated by GW4 and the UK Met Office, funded by EPSRC (EP/P020224/1); HPC Zoo, a research cluster managed by the HPC Group at the University of Bristol (https://uob-hpc.github.io/zoo); Intel DevCloud, an online cluster for developers (https://devcloud.intel.com);EC2 Graviton instances, with access supported by AWS. Publisher Copyright: {\textcopyright} 2022 IEEE.; 2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS), PMBS22 ; Conference date: 14-12-2022",

year = "2023",

month = jan,

day = "30",

doi = "10.1109/PMBS56514.2022.00009",

language = "English",

isbn = "9781665451864",

pages = "36--47",

booktitle = "2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)",

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

address = "United States",

url = "https://www.dcs.warwick.ac.uk/pmbs/pmbs/PMBS/Welcome.html",

}

Lin, T , McIntosh-Smith, SN & Deakin, T 2023, Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems. in 2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS). Institute of Electrical and Electronics Engineers (IEEE), pp. 36-47, 2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS), Dallas, Texas, United States, 14/12/22. https://doi.org/10.1109/PMBS56514.2022.00009

Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems. / Lin, Tom ; McIntosh-Smith, Simon N; Deakin, Tom.
2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS). Institute of Electrical and Electronics Engineers (IEEE), 2023. p. 36-47.

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

TY - GEN

T1 - Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems

AU - Lin, Tom

AU - McIntosh-Smith, Simon N

AU - Deakin, Tom

N1 - Funding Information: ACKNOWLEDGMENT This work makes use of the following services: Isambard UK National Tier-2 HPC Service (https://gw4.ac.uk/isambard) operated by GW4 and the UK Met Office, funded by EPSRC (EP/P020224/1); HPC Zoo, a research cluster managed by the HPC Group at the University of Bristol (https://uob-hpc.github.io/zoo); Intel DevCloud, an online cluster for developers (https://devcloud.intel.com);EC2 Graviton instances, with access supported by AWS. Publisher Copyright: © 2022 IEEE.

PY - 2023/1/30

Y1 - 2023/1/30

N2 - Recent revisions to the ISO C++ standard have added specifications for parallel algorithms. These additions cover common use-cases, including sequence traversal, reduction, and even sorting, many of which are highly applicable in HPC, and thus represent a potential for increased performance and productivity. This study evaluates the state of the art for implementing heterogeneous HPC applications using the latest built-in ISO C++17 parallel algorithms. We implement C++17 ports of representative HPC mini-apps that cover both compute-bound and memory bandwidth-bound applications. We then conduct benchmarks on CPUs and GPUs, comparing our ports to other widely-available parallel programming models, such as OpenMP, CUDA, and SYCL. Finally, we show that C++17 parallel algorithms are able to achieve competitive performance across multiple mini-apps on many platforms, with some notable exceptions. We also discuss several key topics, including portability, and describe workarounds for a number of remaining issues, including index-based traversal and accelerator device/memory management.

AB - Recent revisions to the ISO C++ standard have added specifications for parallel algorithms. These additions cover common use-cases, including sequence traversal, reduction, and even sorting, many of which are highly applicable in HPC, and thus represent a potential for increased performance and productivity. This study evaluates the state of the art for implementing heterogeneous HPC applications using the latest built-in ISO C++17 parallel algorithms. We implement C++17 ports of representative HPC mini-apps that cover both compute-bound and memory bandwidth-bound applications. We then conduct benchmarks on CPUs and GPUs, comparing our ports to other widely-available parallel programming models, such as OpenMP, CUDA, and SYCL. Finally, we show that C++17 parallel algorithms are able to achieve competitive performance across multiple mini-apps on many platforms, with some notable exceptions. We also discuss several key topics, including portability, and describe workarounds for a number of remaining issues, including index-based traversal and accelerator device/memory management.

KW - Performance Portability

KW - Programming Models

KW - GPUs

KW - C++17

KW - PSTL

KW - stdpar

U2 - 10.1109/PMBS56514.2022.00009

DO - 10.1109/PMBS56514.2022.00009

M3 - Conference Contribution (Conference Proceeding)

SN - 9781665451864

SP - 36

EP - 47

BT - 2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)

PB - Institute of Electrical and Electronics Engineers (IEEE)

T2 - 2022 IEEE/ACM International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computer Systems (PMBS)

Y2 - 14 December 2022

ER -

Evaluating ISO C++ Parallel Algorithms on Heterogeneous HPC Systems

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Handle.net

Fingerprint

Cite this