Rocket: efficient and scalable all-pairs computations on heterogeneous platforms

Stijn Heldens, Hein Pieter Hijma, Ben van Werkhoven, Jason Maassen, Henri Bal, Rob van Nieuwpoort

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

19 Downloads (Pure)

Abstract

All-pairs compute problems apply a user-defined function to each combination of two items of a given data set. Although these problems present an abundance of parallelism, data reuse must be exploited to achieve good performance. Several researchers considered this problem, either resorting to partial replication with static work distribution or dynamic scheduling with full replication. In contrast, we present a solution that relies on hierarchical multi-level software-based caches to maximize data reuse at each level in the distributed memory hierarchy, combined with a divide-and-conquer approach to exploit data locality, hierarchical work-stealing to dynamically balance the workload, and asynchronous processing to maximize resource utilization. We evaluate our solution using three real-world applications (from digital forensics, localization microscopy, and bioinformatics) on different platforms (from a desktop machine to a supercomputer). Results shows excellent efficiency and scalability when scaling to 96 GPUs, even obtaining super-linear speedups due to a distributed cache.
Original languageEnglish
Title of host publicationSC20: International Conference for High Performance Computing, Networking, Storage and Analysis
Subtitle of host publication[Proceedings]
PublisherIEEE Computer Society
Number of pages12
ISBN (Electronic)9781728199986
ISBN (Print)9781728199993
DOIs
Publication statusPublished - 2021
Event2020 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2020 - Virtual, Atlanta, United States
Duration: 9 Nov 202019 Nov 2020

Publication series

NameInternational Conference for High Performance Computing, Networking, Storage and Analysis
PublisherIEEE
NumberNovember
Volume2020
ISSN (Print)2167-4329
ISSN (Electronic)2167-4337

Conference

Conference2020 International Conference for High Performance Computing, Networking, Storage and Analysis, SC 2020
Country/TerritoryUnited States
CityVirtual, Atlanta
Period9/11/2019/11/20

Bibliographical note

Publisher Copyright:
© 2020 IEEE.

Keywords

  • all-pairs computation
  • data reuse
  • distributed cache
  • GPU
  • heterogeneous computing
  • work-stealing

Fingerprint

Dive into the research topics of 'Rocket: efficient and scalable all-pairs computations on heterogeneous platforms'. Together they form a unique fingerprint.

Cite this