Accelerator for Sparse Machine Learning

Leonid Yavits; Ran Ginosar

doi:10.1109/lca.2017.2714667

Accelerator for Sparse Machine Learning

Leonid Yavits
, Ran Ginosar

Technion-Israel Institute of Technology

Research output: Contribution to journal › Article › peer-review

15 Scopus citations

Abstract

Sparse matrix by vector multiplication (SpMV) plays a pivotal role in machine learning and data mining. We propose and investigate an SpMV accelerator, specifically designed to accelerate the sparse matrix by sparse vector multiplication (SpMSpV), and to be integrated in a CPU core. We show that our accelerator outperforms a similar solution by 70x while achieving 8x higher power efficiency, which yields an estimated 29x energy reduction for SpMSpV based applications.

Original language	English
Pages (from-to)	21-24
Number of pages	4
Journal	IEEE Computer Architecture Letters
Volume	17
Issue number	1
DOIs	https://doi.org/10.1109/lca.2017.2714667
State	Published - 1 Jan 2018
Externally published	Yes

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Keywords

SpMV
Sparse matrix multiplication
accelerator
sparse matrix by sparse vector multiplication

Access to Document

10.1109/lca.2017.2714667

Cite this

@article{f4cad4281c6149388188a367552119a8,

title = "Accelerator for Sparse Machine Learning",

abstract = "Sparse matrix by vector multiplication (SpMV) plays a pivotal role in machine learning and data mining. We propose and investigate an SpMV accelerator, specifically designed to accelerate the sparse matrix by sparse vector multiplication (SpMSpV), and to be integrated in a CPU core. We show that our accelerator outperforms a similar solution by 70x while achieving 8x higher power efficiency, which yields an estimated 29x energy reduction for SpMSpV based applications.",

keywords = "SpMV, Sparse matrix multiplication, accelerator, sparse matrix by sparse vector multiplication",

author = "Leonid Yavits and Ran Ginosar",

note = "Publisher Copyright: {\textcopyright} 2017 IEEE.",

year = "2018",

month = jan,

day = "1",

doi = "10.1109/lca.2017.2714667",

language = "אנגלית",

volume = "17",

pages = "21--24",

journal = "IEEE Computer Architecture Letters",

issn = "1556-6056",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

number = "1",

}

TY - JOUR

T1 - Accelerator for Sparse Machine Learning

AU - Yavits, Leonid

AU - Ginosar, Ran

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Sparse matrix by vector multiplication (SpMV) plays a pivotal role in machine learning and data mining. We propose and investigate an SpMV accelerator, specifically designed to accelerate the sparse matrix by sparse vector multiplication (SpMSpV), and to be integrated in a CPU core. We show that our accelerator outperforms a similar solution by 70x while achieving 8x higher power efficiency, which yields an estimated 29x energy reduction for SpMSpV based applications.

AB - Sparse matrix by vector multiplication (SpMV) plays a pivotal role in machine learning and data mining. We propose and investigate an SpMV accelerator, specifically designed to accelerate the sparse matrix by sparse vector multiplication (SpMSpV), and to be integrated in a CPU core. We show that our accelerator outperforms a similar solution by 70x while achieving 8x higher power efficiency, which yields an estimated 29x energy reduction for SpMSpV based applications.

KW - SpMV

KW - Sparse matrix multiplication

KW - accelerator

KW - sparse matrix by sparse vector multiplication

UR - https://www.scopus.com/pages/publications/85021800600

U2 - 10.1109/lca.2017.2714667

DO - 10.1109/lca.2017.2714667

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

AN - SCOPUS:85021800600

SN - 1556-6056

VL - 17

SP - 21

EP - 24

JO - IEEE Computer Architecture Letters

JF - IEEE Computer Architecture Letters

IS - 1

ER -

Accelerator for Sparse Machine Learning

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this