Evaluating Performance Portability of OpenMP for SNAP on NVIDIA, Intel, and AMD GPUs Using the Roofline Methodology

被引:1
|
作者
Mehta, Neil A. [1 ]
Gayatri, Rahulkumar [1 ]
Ghadar, Yasaman [2 ]
Knight, Christopher [2 ]
Deslippe, Jack [1 ]
机构
[1] Lawrence Berkeley Natl Lab, NERSC, Berkeley, CA 94720 USA
[2] Argonne Natl Lab, Lemont, IL USA
关键词
Roofline analysis; Performance portability; SNAP; MODEL;
D O I
10.1007/978-3-030-74224-9_1
中图分类号
TP31 [计算机软件];
学科分类号
081202 ; 0835 ;
摘要
In this paper, we show that OpenMP 4.5 based implementation of TestSNAP, a proxy-app for the Spectral Neighbor Analysis Potential (SNAP) in LAMMPS, can be ported across the NVIDIA, Intel, and AMD GPUs. Roofline analysis is employed to assess the performance of TestSNAP on each of the architectures. The main contributions of this paper are two-fold: 1) Provide OpenMP as a viable option for application portability across multiple GPU architectures, and 2) provide a methodology based on the roofline analysis to determine the performance portability of OpenMP implementations on the target architectures. The GPUs used for this work are Intel Gen9, AMD Radeon Instinct MI60, and NVIDIA Volta V100.
引用
收藏
页码:3 / 24
页数:22
相关论文
共 28 条
  • [1] Evaluation of Performance Portability of Applications and Mini-Apps across AMD, Intel and NVIDIA GPUs
    Kwack, JaeHyuk
    Tramm, John
    Bertoni, Colleen
    Ghadar, Yasaman
    Homerding, Brian
    Rangel, Esteban
    Knight, Christopher
    Parker, Scott
    PROCEEDINGS OF 2021 INTERNATIONAL WORKSHOP ON PERFORMANCE, PORTABILITY & PRODUCTIVITY IN HPC (P3HPC 2021), 2021, : 45 - 56
  • [2] Comparing Performance and Portability between CUDA and SYCL for Protein Database Search on NVIDIA, AMD, and Intel GPUs
    Costanzo, Manuel
    Rucci, Enzo
    Garcia-Sanchez, Carlos
    Naiouf, Marcelo
    Prieto-Matias, Manuel
    2023 IEEE 35TH INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING, SBAC-PAD, 2023, : 141 - 148
  • [3] The Productivity, Portability and Performance of OpenMP 4.5 for Scientific Applications Targeting Intel CPUs, IBM CPUs, and NVIDIA GPUs
    Martineau, Matt
    McIntosh-Smith, Simon
    SCALING OPENMP FOR EXASCALE PERFORMANCE AND PORTABILITY (IWOMP 2017), 2017, 10468 : 185 - 200
  • [4] An Empirical Roofline Methodology for Quantitatively Assessing Performance Portability
    Yang, Charlene
    Gayatri, Rahulkumar
    Kurth, Thorsten
    Basu, Protonu
    Ronaghi, Zahra
    Adetokunbo, Adedoyin
    Friesen, Brian
    Cook, Brandon
    Doerfler, Douglas
    Oliker, Leonid
    Deslippe, Jack
    Williams, Samuel
    PROCEEDINGS OF 2018 IEEE/ACM INTERNATIONAL WORKSHOP ON PERFORMANCE, PORTABILITY AND PRODUCTIVITY IN HPC (P3HPC 2018), 2018, : 14 - 23
  • [5] Case Study of Using Kokkos and SYCL as Performance-Portable Frameworks for Milc-Dslash Benchmark on NVIDIA, AMD and Intel GPUs
    Dufek, Amanda S.
    Gayatri, Rahulkumar
    Mehta, Neil
    Doerfler, Douglas
    Cook, Brandon
    Ghadar, Yasaman
    DeTar, Carleton
    PROCEEDINGS OF 2021 INTERNATIONAL WORKSHOP ON PERFORMANCE, PORTABILITY & PRODUCTIVITY IN HPC (P3HPC 2021), 2021, : 57 - 67
  • [6] Performance Portability Evaluation of OpenCL Benchmarks across Intel and NVIDIA Platforms
    Bertoni, Colleen
    Kwack, JaeHyuk
    Applencourt, Thomas
    Ghadar, Yasarnan
    Honierding, Brian
    Knight, Christopher
    Videau, Brice
    Zheng, Huihuo
    Morozov, Vitali
    Parker, Scott
    2020 IEEE 34TH INTERNATIONAL PARALLEL AND DISTRIBUTED PROCESSING SYMPOSIUM WORKSHOPS (IPDPSW 2020), 2020, : 330 - 339
  • [7] Performance Study of an MRI Motion-Compensated Reconstruction Program on Intel CPUs, AMD EPYC CPUs, and NVIDIA GPUs
    Zeroual, Mohamed Aziz
    Isaieva, Karyna
    Vuissoz, Pierre-Andre
    Odille, Freddy
    APPLIED SCIENCES-BASEL, 2024, 14 (21):
  • [8] Evaluating Emerging AI/ML Accelerators: IPU, RDU, and NVIDIA/AMD GPUs
    Peng, Hongwu
    Ding, Caiwen
    Geng, Tong
    Choudhury, Sutanay
    Barker, Kevin
    Li, Ang
    COMPANION OF THE 15TH ACM/SPEC INTERNATIONAL CONFERENCE ON PERFORMANCE ENGINEERING, ICPE COMPANION 2024, 2024, : 14 - 20
  • [9] Performance Assessment of OpenMP Compilers Targeting NVIDIA V100 GPUs
    Davis, Joshua Hoke
    Daley, Christopher
    Pophale, Swaroop
    Huber, Thomas
    Chandrasekaran, Sunita
    Wright, Nicholas J.
    ACCELERATOR PROGRAMMING USING DIRECTIVES, WACCPD 2020, 2021, 12655 : 25 - 44
  • [10] Analysis OpenMP performance of AMD and Intel architecture for breaking waves simulation using MPS
    Alamsyah, M. N. A.
    Utomo, A.
    Gunawan, P. H.
    INTERNATIONAL CONFERENCE ON DATA AND INFORMATION SCIENCE (ICODIS), 2018, 971