The minos computing library: Efficient parallel programming for extremely heterogeneous systems

Roberto Gioiosa, Burcu O. Mutlu, Seyong Lee, Jeffrey S. Vetter, Giulio Picierro, Marco Cesati

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

12 Scopus citations

Abstract

Hardware specialization has become the silver bullet to achieve efficient high performance, from Systems-on-Chip systems, where hardware specialization can be "extreme", to large-scale HPC systems. As the complexity of the systems increases, so does the complexity of programming such architectures in a portable way. This work introduces the Minos Computing Library (MCL), as system software, programming model, and programming model runtime that facilitate programming extremely heterogeneous systems. MCL supports the execution of several multi-threaded applications within the same compute node, performs asynchronous execution of application tasks, efficiently balances computation across hardware resources, and provides performance portability. We show that code developed on a personal desktop automatically scales up to fully utilize powerful workstations with 8 GPUs and down to power-efficient embedded systems. MCL provides up to 17.5x speedup over OpenCL on NVIDIA DGX-1 systems and up to 1.88x speedup on single-GPU systems. In multi-application workloads, MCL's dynamic resource allocation provides up to 2.43x performance improvement over manual, static resources allocation.

Original languageEnglish
Title of host publicationGPGPU 2020 - Proceedings of the 2020 General Purpose Processing Using GPU
PublisherAssociation for Computing Machinery, Inc
Pages1-10
Number of pages10
ISBN (Electronic)9781450370257
DOIs
StatePublished - Feb 23 2020
Event13th Annual Workshop on General Purpose Processing using Graphics Processing Unit, GPGPU 2020 - San Diego, United States
Duration: Feb 23 2020Feb 23 2020

Publication series

NameGPGPU 2020 - Proceedings of the 2020 General Purpose Processing Using GPU

Conference

Conference13th Annual Workshop on General Purpose Processing using Graphics Processing Unit, GPGPU 2020
Country/TerritoryUnited States
CitySan Diego
Period02/23/2002/23/20

Funding

This research is supported by the Department of Energy (DOE) Advanced Scientific Computing Research (ASCR) LAB 19-2119 and Defense Advanced Research Projects Agency (DARPA) HR001117S0055, Program Area Domain-Specific System on Chip (DSSoC).

FundersFunder number
U.S. Department of Energy
Defense Advanced Research Projects AgencyHR001117S0055
Advanced Scientific Computing ResearchLAB 19-2119

    Keywords

    • Asynchronous runtime
    • GPU
    • Heterogeneous systems
    • System software
    • Task-based runtime

    Fingerprint

    Dive into the research topics of 'The minos computing library: Efficient parallel programming for extremely heterogeneous systems'. Together they form a unique fingerprint.

    Cite this