CGPA: Coarse-grained pipelined accelerators

Feng Liu, Soumyadeep Ghosh, Nick P. Johnson, David I. August

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

20 Scopus citations

Abstract

High-level synthesis (HLS) tools dramatically reduce the nonrecurring engineering cost of creating specialized hardware accelerators. Existing HLS tools are successful in synthesizing efficient accelerators for program kernels with regular memory accesses and simple control flows. For other programs, however, these tools yield poor performance because they invoke computation units for instructions sequentially, without exploiting parallelism. To address this problem, this paper proposes Coarse-Grained Pipelined Accelerators (CGPA), an HLS framework that utilizes coarsegrained pipeline parallelism techniques to synthesize efficient specialized accelerator modules from irregular C/C++ programs without requiring any annotations. Compared to the sequential method, CGPA shows speedups of 3.0x-3.8x for 5 kernels from programs in different domains.

Original languageEnglish (US)
Title of host publicationDAC 2014 - 51st Design Automation Conference, Conference Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Print)9781479930173
DOIs
StatePublished - 2014
Event51st Annual Design Automation Conference, DAC 2014 - San Francisco, CA, United States
Duration: Jun 2 2014Jun 5 2014

Publication series

NameProceedings - Design Automation Conference
ISSN (Print)0738-100X

Other

Other51st Annual Design Automation Conference, DAC 2014
Country/TerritoryUnited States
CitySan Francisco, CA
Period6/2/146/5/14

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Modeling and Simulation

Fingerprint

Dive into the research topics of 'CGPA: Coarse-grained pipelined accelerators'. Together they form a unique fingerprint.

Cite this