Efficient GPGPU computing with cross-core resource sharing and core reconfiguration

Ashutosh Dhar, Deming Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

GPUs are capable of running a variety of applications, however their generic parallel-architecture can lead to inefficient use of resources and reduced power efficiency, due to algorithmic or architectural constraints. In this work, taking inspiration from CGRAs (coarse-grained reconfigurable architectures), we demonstrate resource sharing and re-distribution as a solution that can be leveraged by reconfiguring the GPU on a kernel-by-kernel basis. We explore four different schemes that trade the number of active SMs (streaming multiprocessor) for increased occupancy and local memory resources per SM and demonstrate improved power and energy with limited impact to performance. Our most aggressive scheme, BigSM, is capable of saving energy by up to 54%, and 26% on an average.

Original languageEnglish (US)
Title of host publicationProceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages48-55
Number of pages8
ISBN (Electronic)9781538640364
DOIs
StatePublished - Jun 30 2017
Event25th Annual IEEE International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017 - Napa, United States
Duration: Apr 30 2017May 2 2017

Publication series

NameProceedings - IEEE 25th Annual International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017

Other

Other25th Annual IEEE International Symposium on Field-Programmable Custom Computing Machines, FCCM 2017
Country/TerritoryUnited States
CityNapa
Period4/30/175/2/17

Keywords

  • CGRA
  • GPGPU
  • Reconfigurable architecture

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Software
  • Hardware and Architecture
  • Computer Science Applications

Fingerprint

Dive into the research topics of 'Efficient GPGPU computing with cross-core resource sharing and core reconfiguration'. Together they form a unique fingerprint.

Cite this