Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs

Zheming Jin; Jeffrey S. Vetter

doi:10.1109/IPDPSW59300.2023.00067

Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs

Zheming Jin, Jeffrey S. Vetter

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

2 Scopus citations

Abstract

SYCL is a portable programming model. Toward the goal of a better understanding of performance portability of SYCL kernels on GPUs, we select a bioinformatics kernel for computing the all-pairs distance as a case study. After migrating the kernel from CUDA to HIP and SYCL, we evaluate the performance of the CUDA, HIP, and SYCL kernels on NVIDIA V100 and AMD MI210 GPUs. We analyze the GPU instructions from the kernels to explain performance gaps between SYCL and CUDA/HIP. We hope that the findings are valuable for improving performance portability of SYCL.

Original language	English
Title of host publication	2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	366-372
Number of pages	7
ISBN (Electronic)	9798350311990
DOIs	https://doi.org/10.1109/IPDPSW59300.2023.00067
State	Published - 2023
Event	2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023 - St. Petersburg, United States Duration: May 15 2023 → May 19 2023

Publication series

Name	2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023

Conference

Conference	2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023
Country/Territory	United States
City	St. Petersburg
Period	05/15/23 → 05/19/23

Funding

We appreciate the reviewers for their comments and suggestions. The research used resources at the Experimental Computing Lab and the Compute and Data Environment for Science 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. We appreciate the reviewers for their comments and suggestions. The research used resources at the Experimental Computing Lab and the Compute and Data Environment for Science 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.

Keywords

Heterogeneous computing
Performance portabuity
programming model

Access to Document

10.1109/IPDPSW59300.2023.00067

Cite this

Jin, Z., & Vetter, J. S. (2023). Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs. In 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023 (pp. 366-372). (2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IPDPSW59300.2023.00067

Jin, Zheming ; Vetter, Jeffrey S. / Understanding Performance Portability of SYCL Kernels : A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs. 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023. Institute of Electrical and Electronics Engineers Inc., 2023. pp. 366-372 (2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023).

@inproceedings{abc13631adf74a40b1bf273e5f6609dc,

title = "Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs",

abstract = "SYCL is a portable programming model. Toward the goal of a better understanding of performance portability of SYCL kernels on GPUs, we select a bioinformatics kernel for computing the all-pairs distance as a case study. After migrating the kernel from CUDA to HIP and SYCL, we evaluate the performance of the CUDA, HIP, and SYCL kernels on NVIDIA V100 and AMD MI210 GPUs. We analyze the GPU instructions from the kernels to explain performance gaps between SYCL and CUDA/HIP. We hope that the findings are valuable for improving performance portability of SYCL.",

keywords = "Heterogeneous computing, Performance portabuity, programming model",

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

note = "Publisher Copyright: {\textcopyright} 2023 IEEE.; 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023 ; Conference date: 15-05-2023 Through 19-05-2023",

year = "2023",

doi = "10.1109/IPDPSW59300.2023.00067",

language = "English",

series = "2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "366--372",

booktitle = "2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023",

}

Jin, Z & Vetter, JS 2023, Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs. in 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023. 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023, Institute of Electrical and Electronics Engineers Inc., pp. 366-372, 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023, St. Petersburg, United States, 05/15/23. https://doi.org/10.1109/IPDPSW59300.2023.00067

Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs. / Jin, Zheming; Vetter, Jeffrey S.
2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023. Institute of Electrical and Electronics Engineers Inc., 2023. p. 366-372 (2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023).

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

TY - GEN

T1 - Understanding Performance Portability of SYCL Kernels

T2 - 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023

AU - Jin, Zheming

AU - Vetter, Jeffrey S.

PY - 2023

Y1 - 2023

N2 - SYCL is a portable programming model. Toward the goal of a better understanding of performance portability of SYCL kernels on GPUs, we select a bioinformatics kernel for computing the all-pairs distance as a case study. After migrating the kernel from CUDA to HIP and SYCL, we evaluate the performance of the CUDA, HIP, and SYCL kernels on NVIDIA V100 and AMD MI210 GPUs. We analyze the GPU instructions from the kernels to explain performance gaps between SYCL and CUDA/HIP. We hope that the findings are valuable for improving performance portability of SYCL.

AB - SYCL is a portable programming model. Toward the goal of a better understanding of performance portability of SYCL kernels on GPUs, we select a bioinformatics kernel for computing the all-pairs distance as a case study. After migrating the kernel from CUDA to HIP and SYCL, we evaluate the performance of the CUDA, HIP, and SYCL kernels on NVIDIA V100 and AMD MI210 GPUs. We analyze the GPU instructions from the kernels to explain performance gaps between SYCL and CUDA/HIP. We hope that the findings are valuable for improving performance portability of SYCL.

KW - Heterogeneous computing

KW - Performance portabuity

KW - programming model

UR - http://www.scopus.com/inward/record.url?scp=85169292674&partnerID=8YFLogxK

U2 - 10.1109/IPDPSW59300.2023.00067

DO - 10.1109/IPDPSW59300.2023.00067

M3 - Conference contribution

AN - SCOPUS:85169292674

T3 - 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023

SP - 366

EP - 372

BT - 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 15 May 2023 through 19 May 2023

ER -

Jin Z, Vetter JS. Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs. In 2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023. Institute of Electrical and Electronics Engineers Inc. 2023. p. 366-372. (2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023). doi: 10.1109/IPDPSW59300.2023.00067

Understanding Performance Portability of SYCL Kernels: A Case Study with the All-Pairs Distance Calculation in Bioinformatics on GPUs

Abstract

Publication series

Conference

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this