Implementing a GPU programming model on a non-GPU accelerator architecture

Stephen M. Kofsky, Daniel R. Johnson, John A. Stratton, Wen Mei W. Hwu, Sanjay J. Patel, Steven S. Lumetta

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


Parallel codes are written primarily for the purpose of performance. It is highly desirable that parallel codes be portable between parallel architectures without significant performance degradation or code rewrites. While performance portability and its limits have been studied thoroughly on single processor systems, this goal has been less extensively studied and is more difficult to achieve for parallel systems. Emerging single-chip parallel platforms are no exception; writing code that obtains good performance across GPUs and other many-core CMPs can be challenging. In this paper, we focus on CUDA codes, noting that programs must obey a number of constraints to achieve high performance on an NVIDIA GPU. Under such constraints, we develop optimizations that improve the performance of CUDA code on a MIMD accelerator architecture that we are developing called Rigel. We demonstrate performance improvements with these optimizations over naïve translations, and final performance results comparable to those of codes that were hand-optimized for Rigel.

Original languageEnglish (US)
Title of host publicationComputer Architecture - ISCA 2010 International Workshops, A4MMC, AMAS-BT, EAMA, WEED, WIOSCA, Revised Selected Papers
Number of pages12
StatePublished - 2012
EventACM IEEE International Symposium on Computer Architecture, ISCA 2011 - Saint-Malo, France
Duration: Jun 19 2010Jun 23 2010

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume6161 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349


OtherACM IEEE International Symposium on Computer Architecture, ISCA 2011

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Implementing a GPU programming model on a non-GPU accelerator architecture'. Together they form a unique fingerprint.

Cite this