A linear algebra core design for efficient level-3 BLAS

Ardavan Pedram, Syed Zohaib Gilani, Nam Sung Kim, Robert Van De Geijn, Michael Schulte, Andreas Gerstlauer

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

Abstract

Reducing power consumption and increasing effi- ciency is a key concern for many applications. It is well-accepted that specialization and heterogeneity are crucial strategies to improve both power and performance. Yet, how to design highly efficient processing elements while maintaining enough flexibility within a domain of applications is a fundamental question. In this paper, we present the design of a specialized Linear Algebra Core (LAC) for an important class of computational kernels, the level- 3 Basic Linear Algebra Subprograms (BLAS). We demonstrate a detailed algorithm/architecture co-design for mapping a number of level-3 BLAS operations onto the LAC.Results show that our prototype LAC achieves a performance of around 64 GFLOPS (double precision) for these operations, while consuming less than 1.3 Watts in standard 45nm CMOS technology. This is on par with a full-custom design and up to 50x and 10x better in terms of power efficiency than CPUs and GPUs.

Original languageEnglish (US)
Title of host publicationProceedings - 2012 IEEE 23rd International Conference on Application-Specific Systems, Architectures and Processors, ASAP 2012
Pages149-152
Number of pages4
DOIs
StatePublished - Dec 13 2012
Externally publishedYes
Event2012 IEEE 23rd International Conference on Application-Specific Systems, Architectures and Processors, ASAP 2012 - Delft, Netherlands
Duration: Jul 9 2012Jul 11 2012

Publication series

NameProceedings of the International Conference on Application-Specific Systems, Architectures and Processors
ISSN (Print)1063-6862

Other

Other2012 IEEE 23rd International Conference on Application-Specific Systems, Architectures and Processors, ASAP 2012
CountryNetherlands
CityDelft
Period7/9/127/11/12

Keywords

  • BLAS
  • Co-design
  • Linear Algebra
  • Low Power

ASJC Scopus subject areas

  • Hardware and Architecture
  • Computer Networks and Communications

Fingerprint Dive into the research topics of 'A linear algebra core design for efficient level-3 BLAS'. Together they form a unique fingerprint.

Cite this