Microarchitecture-Aware Code Generation for Deep Learning on Single-ISA Heterogeneous Multi-Core Mobile Processors
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Park, Junmo | - |
dc.contributor.author | Kwon, Yongin | - |
dc.contributor.author | Park, Yongjun | - |
dc.contributor.author | Jeon, Dongsuk | - |
dc.date.accessioned | 2022-07-09T19:24:43Z | - |
dc.date.available | 2022-07-09T19:24:43Z | - |
dc.date.created | 2021-05-12 | - |
dc.date.issued | 2019-04 | - |
dc.identifier.issn | 2169-3536 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/148008 | - |
dc.description.abstract | While single-ISA heterogeneous multi-core processors are widely used in mobile computing, typical code generations optimize the code for a single target core, leaving it less suitable for the other cores in the processor. We present a microarchitecture-aware code generation methodology to mitigate this issue. We first suggest adopting Function-Multi-Versioning (FMV) to execute application codes utilizing a core at full capacity regardless of its microarchitecture. We also propose to add a simple but powerful backend optimization pass in the compiler to further boost the performance of applicable cores. Based on these schemes, we developed an automated flow that analyzes the program and generates multiple versions of hot functions tailored to different microarchitectures. At runtime, the running core chooses an optimal version to maximize computation performance. The measurements confirm that the methodology improves the performance of Cortex-A55 and Cortex-A75 cores in Samsung's next-generation Exynos 9820 processor by 11.2% and 17.9%, respectively, while running TensorFlow Lite. | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC | - |
dc.title | Microarchitecture-Aware Code Generation for Deep Learning on Single-ISA Heterogeneous Multi-Core Mobile Processors | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Park, Yongjun | - |
dc.identifier.doi | 10.1109/ACCESS.2019.2910559 | - |
dc.identifier.scopusid | 2-s2.0-85066961444 | - |
dc.identifier.wosid | 000466586900001 | - |
dc.identifier.bibliographicCitation | IEEE ACCESS, v.7, pp.52371 - 52378 | - |
dc.relation.isPartOf | IEEE ACCESS | - |
dc.citation.title | IEEE ACCESS | - |
dc.citation.volume | 7 | - |
dc.citation.startPage | 52371 | - |
dc.citation.endPage | 52378 | - |
dc.type.rims | ART | - |
dc.type.docType | Article | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | Y | - |
dc.description.journalRegisteredClass | scie | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalResearchArea | Engineering | - |
dc.relation.journalResearchArea | Telecommunications | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Information Systems | - |
dc.relation.journalWebOfScienceCategory | Engineering, Electrical & Electronic | - |
dc.relation.journalWebOfScienceCategory | Telecommunications | - |
dc.subject.keywordPlus | Codes (symbols) | - |
dc.subject.keywordPlus | Computer architecture | - |
dc.subject.keywordPlus | Deep learning | - |
dc.subject.keywordPlus | Edge computing | - |
dc.subject.keywordPlus | Application codes | - |
dc.subject.keywordPlus | Automated flow | - |
dc.subject.keywordPlus | Code Generation | - |
dc.subject.keywordPlus | Computation performance | - |
dc.subject.keywordPlus | Micro architectures | - |
dc.subject.keywordPlus | Mobile processors | - |
dc.subject.keywordPlus | Multi-versioning | - |
dc.subject.keywordPlus | Single ISA heterogeneous multi cores | - |
dc.subject.keywordPlus | Multicore programming | - |
dc.subject.keywordAuthor | Edge computing | - |
dc.subject.keywordAuthor | function multi-versioning | - |
dc.subject.keywordAuthor | single-ISA heterogeneous multi-core | - |
dc.identifier.url | https://ieeexplore.ieee.org/document/8688418 | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
222, Wangsimni-ro, Seongdong-gu, Seoul, 04763, Korea+82-2-2220-1365
COPYRIGHT © 2021 HANYANG UNIVERSITY.
Certain data included herein are derived from the © Web of Science of Clarivate Analytics. All rights reserved.
You may not copy or re-distribute this material in whole or in part without the prior written consent of Clarivate Analytics.