DC Field | Value | Language |
---|---|---|
dc.contributor.author | Kim, Joon Kyung | ko |
dc.contributor.author | Ahn, Byung Hoon | ko |
dc.contributor.author | Kinzer, Sean | ko |
dc.contributor.author | Ghodrati, Soroush | ko |
dc.contributor.author | Mahapatra, Rohan | ko |
dc.contributor.author | Yatham, Brahmendra | ko |
dc.contributor.author | Wang, Shu-Ting | ko |
dc.contributor.author | Kim, Dohee | ko |
dc.contributor.author | Sarikhani, Parisa | ko |
dc.contributor.author | Mahmoudi, Babak | ko |
dc.contributor.author | Mahajan, Divya | ko |
dc.contributor.author | Park, Jongse | ko |
dc.contributor.author | Esmaeilzadeh, Hadi | ko |
dc.date.accessioned | 2022-09-27T02:00:34Z | - |
dc.date.available | 2022-09-27T02:00:34Z | - |
dc.date.created | 2022-09-27 | - |
dc.date.created | 2022-09-27 | - |
dc.date.created | 2022-09-27 | - |
dc.date.issued | 2022-09 | - |
dc.identifier.citation | IEEE MICRO, v.42, no.5, pp.89 - 98 | - |
dc.identifier.issn | 0272-1732 | - |
dc.identifier.uri | http://hdl.handle.net/10203/298696 | - |
dc.description.abstract | Field-programmable gate array (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 naturallymakes Cross-Domain Multi-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 developed a dataflow virtual machine, dubbed Accelerator-Level Virtual Machine (XLVM), which transparently maps domain functions (Yin) to best-fit accelerator capabilities (Yang). With six real-world cross-domain applications, our evaluations showthat Yin-Yang unlocks 29.4x speedup, while the best single-domain acceleration achieves 12.0x. | - |
dc.language | English | - |
dc.publisher | IEEE COMPUTER SOC | - |
dc.title | Yin-Yang: Programming Abstractions for Cross-Domain Multi-Acceleration | - |
dc.type | Article | - |
dc.identifier.wosid | 000853841200009 | - |
dc.identifier.scopusid | 2-s2.0-85135735241 | - |
dc.type.rims | ART | - |
dc.citation.volume | 42 | - |
dc.citation.issue | 5 | - |
dc.citation.beginningpage | 89 | - |
dc.citation.endingpage | 98 | - |
dc.citation.publicationname | IEEE MICRO | - |
dc.identifier.doi | 10.1109/MM.2022.3189416 | - |
dc.contributor.localauthor | Park, Jongse | - |
dc.contributor.nonIdAuthor | Kim, Joon Kyung | - |
dc.contributor.nonIdAuthor | Ahn, Byung Hoon | - |
dc.contributor.nonIdAuthor | Kinzer, Sean | - |
dc.contributor.nonIdAuthor | Ghodrati, Soroush | - |
dc.contributor.nonIdAuthor | Mahapatra, Rohan | - |
dc.contributor.nonIdAuthor | Yatham, Brahmendra | - |
dc.contributor.nonIdAuthor | Wang, Shu-Ting | - |
dc.contributor.nonIdAuthor | Kim, Dohee | - |
dc.contributor.nonIdAuthor | Sarikhani, Parisa | - |
dc.contributor.nonIdAuthor | Mahmoudi, Babak | - |
dc.contributor.nonIdAuthor | Mahajan, Divya | - |
dc.contributor.nonIdAuthor | Esmaeilzadeh, Hadi | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Compilers | - |
dc.subject.keywordAuthor | Hardware/Software Interfaces | - |
dc.subject.keywordAuthor | Heterogeneous (Hybrid) Systems | - |
dc.subject.keywordAuthor | Reconfigurable Hardware | - |
dc.subject.keywordAuthor | Runtime Environments | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.