IRIS-BLAS: Towards a Performance Portable and Heterogeneous BLAS Library

被引:5
|
作者
Miniskar, Narasinga Rao [1 ]
Monil, Mohammad Alaul Haque [1 ]
Valero-Lara, Pedro [1 ]
Liu, Frank [1 ]
Vetter, Jeffrey S. [1 ]
机构
[1] Oak Ridge Natl Lab, Comp Sci & Math Div, Oak Ridge, TN 37830 USA
来源
2022 IEEE 29TH INTERNATIONAL CONFERENCE ON HIGH PERFORMANCE COMPUTING, DATA, AND ANALYTICS, HIPC | 2022年
关键词
Performance Portable; Heterogeneity; IRIS; BLAS; Tasking;
D O I
10.1109/HiPC56025.2022.00042
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
This paper presents IRIS-BLAS, a novel heterogeneous and performance portable BLAS library.IRIS-BLAS is built on top of the IRIS runtime and multiple vendor and open-source BLAS libraries. It can transparently use all the architectures/devices available in a heterogeneous system, using the appropriate BLAS library based on the task mapping at run time. Thus, IRIS-BLAS is portable across a broad spectrum of architectures and BLAS libraries, alleviating the worry of application developers about modifying the application source code. Even though the emphasis is on portability, IRIS-BLAS provides competitive or even better performance than other state-of-the-art references. Moreover, IRIS-BLAS offers new features such as efficiently using extremely heterogeneous systems composed of multiple GPUs from different hardware vendors.
引用
收藏
页码:256 / 261
页数:6
相关论文
共 50 条
  • [1] CPU-side High Performance BLAS Library Optimization in Heterogeneous HPL Algorithm
    Cai Y.
    Sun C.-G.
    Du Z.-H.
    Liu Z.-X.
    Kang M.-B.
    Li S.-S.
    Ruan Jian Xue Bao/Journal of Software, 2021, 32 (08): : 2289 - 2306
  • [2] Toward a BLAS library truly portable across different accelerator types
    Eduardo Rodriguez-Gutiez
    Ana Moreton-Fernandez
    Arturo Gonzalez-Escribano
    Diego R. Llanos
    The Journal of Supercomputing, 2019, 75 : 7101 - 7124
  • [3] Toward a BLAS library truly portable across different accelerator types
    Rodriguez-Gutiez, Eduardo
    Moreton-Fernandez, Ana
    Gonzalez-Escribano, Arturo
    Llanos, Diego R.
    JOURNAL OF SUPERCOMPUTING, 2019, 75 (11): : 7101 - 7124
  • [4] Towards an efficient use of the BLAS library for multilinear tensor contractions
    Di Napoli, Edoardo
    Fabregat-Traver, Diego
    Quintana-Orti, Gregorio
    Bientinesi, Paolo
    APPLIED MATHEMATICS AND COMPUTATION, 2014, 235 : 454 - 468
  • [5] CLBlast: A Tuned OpenCL BLAS Library
    Nugteren, Cedric
    IWOCL'18: PROCEEDINGS OF THE INTERNATIONAL WORKSHOP ON OPENCL, 2018, : 22 - 31
  • [6] PORTABLE PARALLEL IMPLEMENTATION OF BLAS-3
    AVERBUCH, A
    AMITAI, D
    FRIEDMAN, R
    GABBER, E
    CONCURRENCY-PRACTICE AND EXPERIENCE, 1994, 6 (05): : 411 - 459
  • [7] FT-BLAS: A High Performance BLAS Implementation With Online Fault Tolerance
    Zhai, Yujia
    Giem, Elisabeth
    Fan, Quan
    Zhao, Kai
    Liu, Jinyang
    Chen, Zizhong
    PROCEEDINGS OF THE 2021 ACM INTERNATIONAL CONFERENCE ON SUPERCOMPUTING, ICS 2021, 2021, : 127 - 138
  • [8] Superscalar GEMM-based level 3 BLAS -: The on-going evolution of a portable and high-performance library
    Gustavson, F
    Henriksson, A
    Jonsson, I
    Kågström, B
    Ling, P
    APPLIED PARALLEL COMPUTING: LARGE SCALE SCIENTIFIC AND INDUSTRIAL PROBLEMS, 1998, 1541 : 207 - 215
  • [9] APPLICATIONS OF LEVEL-2 BLAS IN THE NAG LIBRARY
    DUCROZ, JJ
    MAYES, PJD
    WASNIEWSKI, J
    WILSON, S
    PARALLEL COMPUTING, 1988, 8 (1-3) : 345 - 350
  • [10] PERFORMANCE EVALUATION OF BLAS ON THE TRIDENT PROCESSOR
    Soliman, Mostafa I.
    Sedukhin, Stanislav G.
    PARALLEL PROCESSING LETTERS, 2005, 15 (04) : 407 - 414