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

Jin, Zheming; Vetter, Jeffrey

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

Conference · Mon May 01 00:00:00 EDT 2023

DOI:https://doi.org/10.1109/IPDPSW59300.2023.00067· OSTI ID:1996690

Jin, Zheming ^[1]; ^[1]

ORNL

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.

View Conference

Research Organization:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Organization:: USDOE; USDOE Office of Science (SC)

DOE Contract Number:: AC05-00OR22725

OSTI ID:: 1996690

Country of Publication:: United States

Language:: English

References (19)

Data Parallel C++: Mastering DPC++ for Programming of Heterogeneous Systems using C++ and SYCL Reinders, James; Ashbaugh, Ben; Brodman, James https://doi.org/10.1007/978-1-4842-5574-2	book	November 2020
Automatic OpenCL Code Generation for Multi-device Heterogeneous Architectures Li, Pei; Brunet, Elisabeth; Trahay, Francois 2015 44th International Conference on Parallel Processing (ICPP) https://doi.org/10.1109/ICPP.2015.105	conference	September 2015
Evaluating the Performance of the hipSYCL Toolchain for HPC Kernels on NVIDIA V100 GPUs Homerding, Brian; Tramm, John Proceedings of the International Workshop on OpenCL https://doi.org/10.1145/3388333.3388660	conference	April 2020
Performance portability study of epistasis detection using SYCL on NVIDIA GPU Jin, Zheming; Vetter, Jeffrey S. Proceedings of the 13th ACM International Conference on Bioinformatics, Computational Biology and Health Informatics https://doi.org/10.1145/3535508.3545591	conference	August 2022
Bringing performant support for Nvidia® hardware to SYCL Reyes, Ruyman; Brown, Gordon; Burns, Rod Proceedings of the International Workshop on OpenCL https://doi.org/10.1145/3388333.3388651	conference	April 2020
Kokkos 3: Programming Model Extensions for the Exascale Era Trott, Christian; Lebrun-Grandie, Damien; Arndt, Daniel IEEE Transactions on Parallel and Distributed Systems https://doi.org/10.1109/TPDS.2021.3097283	journal	January 2021
Experiences Porting NAMD to the Data Parallel C++ Programming Model Hardy, David J.; Choi, Jaemin; Jiang, Wei International Workshop on OpenCL https://doi.org/10.1145/3529538.3529560	conference	May 2022
Exploiting graphics processing units for computational biology and bioinformatics Payne, Joshua L.; Sinnott-Armstrong, Nicholas A.; Moore, Jason H. Interdisciplinary Sciences: Computational Life Sciences, Vol. 2, Issue 3 https://doi.org/10.1007/s12539-010-0002-4	journal	July 2010
SkelCL: a high-level extension of OpenCL for multi-GPU systems Steuwer, Michel; Gorlatch, Sergei The Journal of Supercomputing, Vol. 69, Issue 1 https://doi.org/10.1007/s11227-014-1213-y	journal	May 2014
Evaluation of Intel's DPC++ Compatibility Tool in heterogeneous computing Castaño, Germán; Faqir-Rhazoui, Youssef; García, Carlos Journal of Parallel and Distributed Computing, Vol. 165 https://doi.org/10.1016/j.jpdc.2022.03.017	journal	July 2022
NVIDIA Tesla: A Unified Graphics and Computing Architecture Lindholm, Erik; Nickolls, John; Oberman, Stuart IEEE Micro, Vol. 28, Issue 2 https://doi.org/10.1109/MM.2008.31	journal	March 2008
Parallel Computing Experiences with CUDA Garland, Michael; Le Grand, Scott; Nickolls, John IEEE Micro, Vol. 28, Issue 4 https://doi.org/10.1109/MM.2008.57	journal	July 2008
Understanding Performance Portability of Bioinformatics Applications in SYCL on an NVIDIA GPU Jin, Zheming; Vetter, Jeffrey S. 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) https://doi.org/10.1109/BIBM55620.2022.9995222	conference	December 2022
Lost in Abstraction: Pitfalls of Analyzing GPUs at the Intermediate Language Level Gutierrez, Anthony; Beckmann, Bradford M.; Dutu, Alexandru 2018 IEEE International Symposium on High Performance Computer Architecture (HPCA) https://doi.org/10.1109/HPCA.2018.00058	conference	February 2018
Evaluating the performance of HPC-style SYCL applications Deakin, Tom; McIntosh-Smith, Simon Proceedings of the International Workshop on OpenCL https://doi.org/10.1145/3388333.3388643	conference	April 2020
Rodinia: A benchmark suite for heterogeneous computing Che, Shuai; Boyer, Michael; Meng, Jiayuan 2009 IEEE International Symposium on Workload Characterization (IISWC) https://doi.org/10.1109/IISWC.2009.5306797	conference	October 2009
The X^e GPU Architecture Blythe, David 2020 IEEE Hot Chips 32 Symposium (HCS) https://doi.org/10.1109/HCS49909.2020.9220591	conference	August 2020
Evaluating Performance and Portability of a core bioinformatics kernel on multiple vendor GPUs Haseeb, Muhammad; Ding, Nan; Deslippe, Jack 2021 International Workshop on Performance, Portability and Productivity in HPC (P3HPC) https://doi.org/10.1109/P3HPC54578.2021.00010	conference	November 2021
Alpaka -- An Abstraction Library for Parallel Kernel Acceleration Zenker, Erik; Worpitz, Benjamin; Widera, Rene 2016 IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) https://doi.org/10.1109/IPDPSW.2016.50	conference	May 2016

Similar Records

Understanding Performance Portability of Bioinformatics Applications in SYCL on an NVIDIA GPU

Conference · Wed Nov 30 23:00:00 EST 2022 · OSTI ID:1909099

Understanding SYCL Portability for Pseudorandom Number Generation: a Case Study with Gene-Expression Connectivity Mapping

Conference · Mon May 01 00:00:00 EDT 2023 · OSTI ID:1996688

Case Study of Using Kokkos and SYCLs Performance-Portable Frameworks for Milc-Dslash Benchmark on NVIDIA, AMD and Intel GPUs

Conference · Thu Dec 31 23:00:00 EST 2020 · OSTI ID:1892057

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

Citation Formats

References (19)

Similar Records

Related Subjects