Decoupling data supply from computation for latency-tolerant communication in heterogeneous architectures

Tae Jun Ham, Juan L. Aragón, Margaret Rose Martonosi

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

In today's computers, heterogeneous processing is used to meet performance targets at manageable power. In adopting increased compute specialization, however, the relative amount of time spent on communication increases. System and software optimizations for communication often come at the costs of increased complexity and reduced portability. The Decoupled Supply-Compute (DeSC) approach offers a way to attack communication latency bottlenecks automatically, while maintaining good portability and low complexity. Our work expands prior Decoupled Access Execute techniques with hardware/software specialization. For a range of workloads, DeSC offers roughly 2× speedup, and additional specialized compression optimizations reduce traffic between decoupled units by 40%.

Original languageEnglish (US)
Article number16
JournalACM Transactions on Architecture and Code Optimization
Volume14
Issue number2
DOIs
StatePublished - Jun 2017

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems
  • Hardware and Architecture

Keywords

  • Accelerators
  • Communication management
  • Decoupled architecture

Fingerprint Dive into the research topics of 'Decoupling data supply from computation for latency-tolerant communication in heterogeneous architectures'. Together they form a unique fingerprint.

  • Cite this