skip to main content
10.1145/3332466.3374530acmconferencesArticle/Chapter ViewAbstractPublication PagesppoppConference Proceedingsconference-collections
poster

Revisiting linpack algorithm on large-scale CPU-GPU heterogeneous systems

Published: 19 February 2020 Publication History

Abstract

As the widening gap between GPU computing capability and other components (CPU, PCIe bus and communication network), it's increasingly challenging to design high performance parallel algorithms for large CPU-GPU heterogeneous systems. There are mainly two reasons. Firstly, simply offloading the kernel library to GPU incurs large volume data transfer through low-speed PCIe bus. Secondly, communication overheads through network severely affects scalability. To solve the above issues, we advocate a paradigm shift to CPU-centric and fine-grained pipelining algorithm design. By taking Linpack benchmark as a case study, the new algorithm design paradigm shows its effectiveness. Our optimized Linpack program achieves 63.79PFlops on 16384 GPUs. Its floating-point efficiency outperforms the NVIDIA proprietary counterparts by 5% on average.

References

[1]
ABCI. 2019. AI Bridging Cloud Infrastructure (ABCI). https://abci.ai/en/about_abci/
[2]
Jack Dongarra, Robert van de Geijn, and David Walker. 1994. Scalability Issues Affecting The Design Of A Dense Linear Algebra Library. J. Parallel and Distrib. Comput. (06 1994).
[3]
G. Jo, J. Nah, J. Lee, J. Kim, and J. Lee. 2015. Accelerating LINPACK with MPI-OpenCL on Clusters of Multi-GPU Nodes. IEEE Transactions on Parallel and Distributed Systems 26, 7 (July 2015), 1814--1825.
[4]
Antoine Petitet, R C. Whaley, Jack Dongarra, and A Cleary. 2008. HPL - a Portable Implementation of the High-Performance Linpack Benchmark for Distributed-Memory Computers. (01 2008).
[5]
TOP500. 2019. TOP500 List, https://www.top500.org/lists/2019/11/

Cited By

View all
  • (2022)Evolving the HPL benchmark towards multi-GPGPU clustersCCF Transactions on High Performance Computing10.1007/s42514-022-00128-65:1(84-96)Online publication date: 26-Oct-2022
  • (2020)A novel approach for radar network’s detection power analysis based on GPU2020 Eighth International Conference on Advanced Cloud and Big Data (CBD)10.1109/CBD51900.2020.00015(31-36)Online publication date: Dec-2020

Index Terms

  1. Revisiting linpack algorithm on large-scale CPU-GPU heterogeneous systems

    Recommendations

    Comments

    Information & Contributors

    Information

    Published In

    cover image ACM Conferences
    PPoPP '20: Proceedings of the 25th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming
    February 2020
    454 pages
    ISBN:9781450368186
    DOI:10.1145/3332466
    Permission to make digital or hard copies of part or all of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. Copyrights for third-party components of this work must be honored. For all other uses, contact the Owner/Author.

    Sponsors

    Publisher

    Association for Computing Machinery

    New York, NY, United States

    Publication History

    Published: 19 February 2020

    Check for updates

    Author Tags

    1. Linpack algorithm
    2. heterogeneous system
    3. software pipeline

    Qualifiers

    • Poster

    Funding Sources

    • Chinese Academy of Sciences
    • National Natural Science Foundation of China
    • The National Key Research and Development Program of China

    Conference

    PPoPP '20

    Acceptance Rates

    PPoPP '20 Paper Acceptance Rate 28 of 121 submissions, 23%;
    Overall Acceptance Rate 230 of 1,014 submissions, 23%

    Upcoming Conference

    Contributors

    Other Metrics

    Bibliometrics & Citations

    Bibliometrics

    Article Metrics

    • Downloads (Last 12 months)27
    • Downloads (Last 6 weeks)1
    Reflects downloads up to 14 Feb 2025

    Other Metrics

    Citations

    Cited By

    View all
    • (2022)Evolving the HPL benchmark towards multi-GPGPU clustersCCF Transactions on High Performance Computing10.1007/s42514-022-00128-65:1(84-96)Online publication date: 26-Oct-2022
    • (2020)A novel approach for radar network’s detection power analysis based on GPU2020 Eighth International Conference on Advanced Cloud and Big Data (CBD)10.1109/CBD51900.2020.00015(31-36)Online publication date: Dec-2020

    View Options

    Login options

    View options

    PDF

    View or Download as a PDF file.

    PDF

    eReader

    View online with eReader.

    eReader

    Figures

    Tables

    Media

    Share

    Share

    Share this Publication link

    Share on social media