Experience Deploying Graph Applications on GPUs with SYCL

Zheming Jin; Jeffrey S. Vetter

doi:10.1145/3605731.3605744

Experience Deploying Graph Applications on GPUs with SYCL

Zheming Jin, Jeffrey S. Vetter

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

2 Scopus citations

Abstract

SYCL allows for deployment and use of accelerators across vendors' platforms. In this work, we describe the experience of deploying graph analytics on vendors' GPUs using SYCL. We contrast the CUDA and SYCL application programming interfaces by describing the experience of migrating the applications from CUDA to SYCL, evaluate the performance of the applications on NVIDIA and AMD GPUs, and explore performance improvement with device-level parallelism. The results show that the recent SYCL extensions facilitate functional portability, but improving code optimizations and resource usage for performance portability is needed in the compiler implementation.

Original language	English
Title of host publication	52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings
Publisher	Association for Computing Machinery
Pages	30-39
Number of pages	10
ISBN (Electronic)	9798400708435
DOIs	https://doi.org/10.1145/3605731.3605744
State	Published - Aug 7 2023
Event	52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings - Salt Lake City, United States Duration: Aug 7 2023 → Aug 10 2023

Publication series

Name	ACM International Conference Proceeding Series

Conference

Conference	52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings
Country/Territory	United States
City	Salt Lake City
Period	08/7/23 → 08/10/23

Funding

We appreciate the reviewers for their comments and ACM TAPS for their assistance. This research used resources of the Experimental Computing Lab at Oak Ridge National Laboratory. This research was supported by the US Department of Energy Advanced Scientific Computing Research program under Contract No. DE-AC05-00OR22725. This research used resources of the Experimental Computing Lab at Oak Ridge National Laboratory. This research was supported by the US Department of Energy Advanced Scientific Computing Research program under Contract No. DEAC05-00OR22725.

Keywords

GPUs
SYCL
portability

Access to Document

10.1145/3605731.3605744

Cite this

@inproceedings{779a4b98a19045b0af9a5b23f900f545,

title = "Experience Deploying Graph Applications on GPUs with SYCL",

abstract = "SYCL allows for deployment and use of accelerators across vendors' platforms. In this work, we describe the experience of deploying graph analytics on vendors' GPUs using SYCL. We contrast the CUDA and SYCL application programming interfaces by describing the experience of migrating the applications from CUDA to SYCL, evaluate the performance of the applications on NVIDIA and AMD GPUs, and explore performance improvement with device-level parallelism. The results show that the recent SYCL extensions facilitate functional portability, but improving code optimizations and resource usage for performance portability is needed in the compiler implementation.",

keywords = "GPUs, SYCL, portability",

author = "Zheming Jin and Vetter, {Jeffrey S.}",

note = "Publisher Copyright: {\textcopyright} 2023 ACM.; 52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings ; Conference date: 07-08-2023 Through 10-08-2023",

year = "2023",

month = aug,

day = "7",

doi = "10.1145/3605731.3605744",

language = "English",

series = "ACM International Conference Proceeding Series",

publisher = "Association for Computing Machinery",

pages = "30--39",

booktitle = "52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings",

}

Jin, Z & Vetter, JS 2023, Experience Deploying Graph Applications on GPUs with SYCL. in 52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings. ACM International Conference Proceeding Series, Association for Computing Machinery, pp. 30-39, 52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings, Salt Lake City, United States, 08/7/23. https://doi.org/10.1145/3605731.3605744

Experience Deploying Graph Applications on GPUs with SYCL. / Jin, Zheming; Vetter, Jeffrey S.
52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings. Association for Computing Machinery, 2023. p. 30-39 (ACM International Conference Proceeding Series).

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

TY - GEN

T1 - Experience Deploying Graph Applications on GPUs with SYCL

AU - Jin, Zheming

AU - Vetter, Jeffrey S.

PY - 2023/8/7

Y1 - 2023/8/7

N2 - SYCL allows for deployment and use of accelerators across vendors' platforms. In this work, we describe the experience of deploying graph analytics on vendors' GPUs using SYCL. We contrast the CUDA and SYCL application programming interfaces by describing the experience of migrating the applications from CUDA to SYCL, evaluate the performance of the applications on NVIDIA and AMD GPUs, and explore performance improvement with device-level parallelism. The results show that the recent SYCL extensions facilitate functional portability, but improving code optimizations and resource usage for performance portability is needed in the compiler implementation.

AB - SYCL allows for deployment and use of accelerators across vendors' platforms. In this work, we describe the experience of deploying graph analytics on vendors' GPUs using SYCL. We contrast the CUDA and SYCL application programming interfaces by describing the experience of migrating the applications from CUDA to SYCL, evaluate the performance of the applications on NVIDIA and AMD GPUs, and explore performance improvement with device-level parallelism. The results show that the recent SYCL extensions facilitate functional portability, but improving code optimizations and resource usage for performance portability is needed in the compiler implementation.

KW - GPUs

KW - SYCL

KW - portability

UR - http://www.scopus.com/inward/record.url?scp=85175001487&partnerID=8YFLogxK

U2 - 10.1145/3605731.3605744

DO - 10.1145/3605731.3605744

M3 - Conference contribution

AN - SCOPUS:85175001487

T3 - ACM International Conference Proceeding Series

SP - 30

EP - 39

BT - 52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings

PB - Association for Computing Machinery

T2 - 52nd International Conference on Parallel Processing, ICPP 2023 - Workshops Proceedings

Y2 - 7 August 2023 through 10 August 2023

ER -

Experience Deploying Graph Applications on GPUs with SYCL

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this