DANBI: Dynamic scheduling of irregular stream programs for many-core systems
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Min C.[Min C.] | - |
dc.contributor.author | Eom Y.I.[Eom Y.I.] | - |
dc.date.accessioned | 2021-08-04T23:46:02Z | - |
dc.date.available | 2021-08-04T23:46:02Z | - |
dc.date.created | 2016-08-06 | - |
dc.date.issued | 2013 | - |
dc.identifier.issn | 1089-795X | - |
dc.identifier.uri | https://scholarworks.bwise.kr/skku/handle/2021.sw.skku/62635 | - |
dc.description.abstract | The stream programming model has received a lot of interest because it naturally exposes task, data, and pipeline parallelism. However, most prior work has focused on static scheduling of regular stream programs. Therefore, irregular applications cannot be handled in static scheduling, and the load imbalance caused by static scheduling faces scalability limitations in many-core systems. In this paper, we introduce the DANBI1 programming model which supports irregular stream programs and propose dynamic scheduling techniques. Scheduling irregular stream programs is very challenging and the load imbalance becomes a major hurdle to achieve scalability. Our dynamic load-balancing scheduler exploits producer-consumer relationships already expressed in the stream program to achieve scalability. Moreover, it effectively avoids the thundering-herd problem and dynamically adapts to load imbalance in a probabilistic manner. It surpasses prior static stream scheduling approaches which are vulnerable to load imbalance and also surpasses prior dynamic stream scheduling approaches which have many restrictions on supported program types, on the scope of dynamic scheduling, and on preserving data ordering. Our experimental results on a 40-core server show that DANBI achieves an almost linear scalability and outperforms state-of-the-art parallel runtimes by up to 2.8 times. © 2013 IEEE. | - |
dc.subject | Dynamic scheduling | - |
dc.subject | Irregular applications | - |
dc.subject | Irregular programs | - |
dc.subject | Pipeline parallelisms | - |
dc.subject | Programming models | - |
dc.subject | Software pipelining | - |
dc.subject | Stream programming | - |
dc.subject | Stream Scheduling | - |
dc.subject | Dynamic loads | - |
dc.subject | Network management | - |
dc.subject | Parallel architectures | - |
dc.subject | Resource allocation | - |
dc.subject | Scalability | - |
dc.subject | Scheduling | - |
dc.title | DANBI: Dynamic scheduling of irregular stream programs for many-core systems | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Min C.[Min C.] | - |
dc.contributor.affiliatedAuthor | Eom Y.I.[Eom Y.I.] | - |
dc.identifier.doi | 10.1109/PACT.2013.6618816 | - |
dc.identifier.scopusid | 2-s2.0-84887438891 | - |
dc.identifier.bibliographicCitation | Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT, pp.189 - 200 | - |
dc.relation.isPartOf | Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT | - |
dc.citation.title | Parallel Architectures and Compilation Techniques - Conference Proceedings, PACT | - |
dc.citation.startPage | 189 | - |
dc.citation.endPage | 200 | - |
dc.type.rims | ART | - |
dc.description.journalClass | 1 | - |
dc.description.journalRegisteredClass | scopus | - |
dc.subject.keywordAuthor | Irregular Programs | - |
dc.subject.keywordAuthor | Load Balancing | - |
dc.subject.keywordAuthor | Scheduling | - |
dc.subject.keywordAuthor | Software Pipelining | - |
dc.subject.keywordAuthor | Stream Programming | - |
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.
(03063) 25-2, SUNGKYUNKWAN-RO, JONGNO-GU, SEOUL, KOREAsamsunglib@skku.edu
COPYRIGHT © 2021 SUNGKYUNKWAN UNIVERSITY ALL RIGHTS RESERVED.
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.