The Kokkos EcoSystem: Comprehensive Performance Portability for High Performance Computing

被引：48

作者：

Trott, Christian ^{[1
]}

Berger-Vergiat, Luc ^{[1
]}

Poliakoff, David ^{[1
]}

Rajamanickam, Sivasankaran ^{[1
]}

Lebrun-Grandie, Damien ^{[2
]}

Madsen, Jonathan ^{[3
]}

Al Awar, Nader ^{[4
]}

Gligoric, Milos ^{[4
]}

Shipman, Galen ^{[5
]}

Womeldorff, Geoff ^{[5
]}

机构：

[1] Sandia Natl Labs, Albuquerque, NM 87123 USA

[2] Oak Ridge Natl Lab, Oak Ridge, TN 37830 USA

[3] Lawrence Berkeley Natl Lab, Natl Energy Res Sci Comp Ctr, Berkeley, CA 94720 USA

[4] Univ Texas Austin, Dept Elect & Comp Engn, Austin, TX 78712 USA

[5] Los Alamos Natl Lab, Los Alamos, NM 87545 USA

来源：

COMPUTING IN SCIENCE & ENGINEERING | 2021年 / 23卷 / 05期

关键词：

High performance computing; Performance evaluation; Programming; Computer architecture; Debugging; Ecosystems;

D O I：

10.1109/MCSE.2021.3098509

中图分类号：

TP39 [计算机的应用];

学科分类号：

081203 ; 0835 ;

摘要：

State-of-the-art engineering and science codes have grown in complexity dramatically over the last two decades. Application teams have adopted more sophisticated development strategies, leveraging third party libraries, deploying comprehensive testing, and using advanced debugging and profiling tools. In today's environment of diverse hardware platforms, these applications also desire performance portability-avoiding the need to duplicate work for various platforms. The Kokkos EcoSystem provides that portable software stack. Based on the Kokkos Core Programming Model, the EcoSystem provides math libraries, interoperability capabilities with Python and Fortran, and Tools for analyzing, debugging, and optimizing applications. In this article, we overview the components, discuss some specific use cases, and highlight how codesigning these components enables a more developer friendly experience.

引用

页码：10 / 18

页数：9

共 6 条

[1] A Performance Portability Framework for Python']Python [J].

Al Awar, Nader ;

Zhu, Steven ;

Biros, George ;

Gligoric, Milos .

PROCEEDINGS OF THE 2021 ACM INTERNATIONAL CONFERENCE ON SUPERCOMPUTING, ICS 2021, 2021, :467-478

[2] HOMMEXX 1.0: a performance-portable atmospheric dynamical core for the Energy Exascale Earth System Model [J].

Bertagna, Luca ;

Deakin, Michael ;

Guba, Oksana ;

Sunderland, Daniel ;

Bradley, Andrew M. ;

Tezaur, Irina K. ;

Taylor, Mark A. ;

Salinger, Andrew G. .

GEOSCIENTIFIC MODEL DEVELOPMENT, 2019, 12 (04) :1423-1441

[3] Kokkos: Enabling manycore performance portability through polymorphic memory access patterns [J].

Edwards, H. Carter ;

Trott, Christian R. ;

Sunderland, Daniel .

JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2014, 74 (12) :3202-3216

[4] Designing Vector-Friendly Compact BLAS and LAPACK Kernels [J].

Kim, Kyungjoo ;

Costa, Timothy B. ;

Deveci, Mehmet ;

Bradley, Andrew M. ;

Hammond, Simon D. ;

Guney, Murat E. ;

Knepper, Sarah ;

Story, Shane ;

Rajamanickam, Sivasankaran .

SC'17: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE FOR HIGH PERFORMANCE COMPUTING, NETWORKING, STORAGE AND ANALYSIS, 2017,

[5] FAST PARALLEL ALGORITHMS FOR SHORT-RANGE MOLECULAR-DYNAMICS [J].

PLIMPTON, S .

JOURNAL OF COMPUTATIONAL PHYSICS, 1995, 117 (01) :1-19

[6]

Rajamanickam S, ARXIV210311991, V2021

← 1 →