IEEE Micro最新文献

筛选
英文 中文
Yin-Yang: Programming Abstractions for Cross-Domain Multi-Acceleration. 尹扬:跨领域多加速程序设计摘要。
IF 3.6 3区 计算机科学
IEEE Micro Pub Date : 2022-09-01 Epub Date: 2022-08-01 DOI: 10.1109/mm.2022.3189416
Joon Kyung Kim, Byung Hoon Ahn, Sean Kinzer, Soroush Ghodrati, Rohan Mahapatra, Brahmendra Yatham, Shu-Ting Wang, Dohee Kim, Parisa Sarikhani, Babak Mahmoudi, Divya Mahajan, Jongse Park, Hadi Esmaeilzadeh
{"title":"Yin-Yang: Programming Abstractions for Cross-Domain Multi-Acceleration.","authors":"Joon Kyung Kim,&nbsp;Byung Hoon Ahn,&nbsp;Sean Kinzer,&nbsp;Soroush Ghodrati,&nbsp;Rohan Mahapatra,&nbsp;Brahmendra Yatham,&nbsp;Shu-Ting Wang,&nbsp;Dohee Kim,&nbsp;Parisa Sarikhani,&nbsp;Babak Mahmoudi,&nbsp;Divya Mahajan,&nbsp;Jongse Park,&nbsp;Hadi Esmaeilzadeh","doi":"10.1109/mm.2022.3189416","DOIUrl":"10.1109/mm.2022.3189416","url":null,"abstract":"<p><p>FPGA accelerators offer performance and efficiency gains by narrowing the scope of acceleration to one algorithmic domain. However, real-life applications are often not limited to a single domain, which naturally makes <i>Cross</i>-Domain <i>Multi</i>-Acceleration a crucial next step. The challenge is, existing FPGA accelerators are built upon their specific vertically-specialized stacks, which prevents utilizing multiple accelerators from different domains. To that end, we propose a pair of dual abstractions, called Yin-Yang, which work in tandem and enable programmers to develop cross-domain applications using multiple accelerators on a FPGA. The Yin abstraction enables cross-domain algorithmic specification, while the Yang abstraction captures the accelerator capabilities. We also develop a dataflow virtual machine, dubbed XLVM, that transparently maps domain functions (Yin) to best-fit accelerator capabilities (Yang). With six real-world cross-domain applications, our evaluations show that Yin-Yang unlocks 29.4× speedup, while the best single-domain acceleration achieves 12.0×.</p>","PeriodicalId":13100,"journal":{"name":"IEEE Micro","volume":"42 5","pages":"89-98"},"PeriodicalIF":3.6,"publicationDate":"2022-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10065469/pdf/nihms-1836431.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"10190559","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
引用次数: 0
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
相关产品
×
本文献相关产品
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信