A Hardware-Software Blueprint for Flexible Deep Learning Specialization

被引：86

作者：

Moreau, Thierry ^{[1
]}

Chen, Tianqi ^{[2
]}

Vega, Luis ^{[2
]}

Roesch, Jared ^{[1
]}

Yan, Eddie ^{[2
]}

Zheng, Lianmin ^{[3
]}

Fromm, Josh ^{[1
]}

Jiang, Ziheng ^{[1
]}

Ceze, Luis ^{[2
]}

Guestrin, Carlos ^{[4
]}

Krishnamurthy, Arvind ^{[5
]}

机构：

[1] Univ Washington, Seattle, WA 98195 USA

[2] Univ Washington, Paul G Allen Sch Comp Sci & Engn Dept, Seattle, WA 98195 USA

[3] Shanghai Jiao Tong Univ, Shanghai, Peoples R China

[4] Univ Washington, Comp Sci & Engn Dept, Machine Learning, Seattle, WA 98195 USA

[5] Univ Washington, Comp Sci & Engn Dept, Seattle, WA 98195 USA

来源：

IEEE MICRO | 2019年 / 39卷 / 05期

基金：

美国国家科学基金会;

关键词：

D O I：

10.1109/MM.2019.2928962

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

This article describes the Versatile Tensor Accelerator (VTA), a programmable DL architecture designed to be extensible in the face of evolving workloads. VTA achieves "flexible specialization" via a parameterizable architecture, two-level Instruction Set Architecture (ISA), and a Just in Time (JIT) compiler.

引用

页码：8 / 16

页数：9

共 8 条

[1] Chen T., 2015, P NEUR INF PROC SYST
[2] Chen TQ, 2018, PROCEEDINGS OF THE 13TH USENIX SYMPOSIUM ON OPERATING SYSTEMS DESIGN AND IMPLEMENTATION, P579
[3] Chen W, 2018, CHIN AUTOM CONGR, P3389, DOI 10.1109/CAC.2018.8623478
[4] Jamieson K, 2016, JMLR WORKSH CONF PRO, V51, P240
[5] Jiang H, 2016, INT C PAR DISTRIB SY, P785, DOI [10.1109/ICPADS.2016.0107, 10.1109/ICPADS.2016.105]
[6] In-Datacenter Performance Analysis of a Tensor Processing Unit
Jouppi, Norman P.
Young, Cliff
Patil, Nishant
Patterson, David
Agrawal, Gaurav
Bajwa, Raminder
Bates, Sarah
Bhatia, Suresh
Boden, Nan
Borchers, Al
Boyle, Rick
Cantin, Pierre-luc
Chao, Clifford
Clark, Chris
Coriell, Jeremy
Daley, Mike
Dau, Matt
Dean, Jeffrey
Gelb, Ben
Ghaemmaghami, Tara Vazir
Gottipati, Rajendra
Gulland, William
Hagmann, Robert
Ho, C. Richard
Hogberg, Doug
Hu, John
Hundt, Robert
Hurt, Dan
Ibarz, Julian
Jaffey, Aaron
Jaworski, Alek
Kaplan, Alexander
Khaitan, Harshit
Killebrew, Daniel
Koch, Andy
Kumar, Naveen
Lacy, Steve
Laudon, James
Law, James
Le, Diemthu
Leary, Chris
Liu, Zhuyuan
Lucke, Kyle
Lundin, Alan
MacKean, Gordon
Maggiore, Adriana
Mahony, Maire
Miller, Kieran
Nagarajan, Rahul
Narayanaswami, Ravi
[J]. 44TH ANNUAL INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE (ISCA 2017), 2017, : 1 - 12
[7] Relay: A New IR for Machine Learning Frameworks
Roesch, Jared
Lyubomirsky, Steven
Weber, Logan
Pollock, Josh
Kirisame, Marisa
Chen, Tianqi
Tatlock, Zachary
[J]. MAPL'18: PROCEEDINGS OF THE 2ND ACM SIGPLAN INTERNATIONAL WORKSHOP ON MACHINE LEARNING AND PROGRAMMING LANGUAGES, 2018, : 58 - 68
[8] Smith J. E., 1982, 9th Annual Symposium on Computer Architecture, P112

← 1 →