Stream Processors: Progammability and Efficiency

Author:

Dally William J.1,Kapasi Ujval J.1,Khailany Brucek1,Ahn Jung Ho1,Das Abhishek1

Affiliation:

1. Stanford University

Abstract

Many signal processing applications require both efficiency and programmability. Baseband signal processing in 3G cellular base stations, for example, requires hundreds of GOPS (giga, or billions, of operations per second) with a power budget of a few watts, an efficiency of about 100 GOPS/W (GOPS per watt), or 10 pJ/op (picoJoules per operation). At the same time programmability is needed to follow evolving standards, to support multiple air interfaces, and to dynamically provision processing resources over different air interfaces. Digital television, surveillance video processing, automated optical inspection, and mobile cameras, camcorders, and 3G cellular handsets have similar needs.

Publisher

Association for Computing Machinery (ACM)

Subject

General Computer Science

Reference10 articles.

1. 4 . A high-end superscalar processor may consume 10 nJ or more per instruction because of much greater overheads required for acceleration techniques such as branch prediction register renaming and out-of-order execution. 4. A high-end superscalar processor may consume 10 nJ or more per instruction because of much greater overheads required for acceleration techniques such as branch prediction register renaming and out-of-order execution.

Cited by 20 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

1. Occam: Optimal Data Reuse for Convolutional Neural Networks;ACM Transactions on Architecture and Code Optimization;2022-12-16

2. Toward Real-Time Ray Tracing;ACM Computing Surveys;2018-07-31

3. UDP;Proceedings of the 50th Annual IEEE/ACM International Symposium on Microarchitecture;2017-10-14

4. A case study of different task implementations for multioutput stages in non-trivial parallel pipeline applications;Parallel Computing;2014-08

5. Scheduling parity checks for increased throughput in early-termination, layered decoding of QC-LDPC codes on a stream processor;EURASIP Journal on Wireless Communications and Networking;2012-04-12

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3