A Specialized Architecture for Object Serialization with Applications to Big Data Analytics
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Jang, J.[Jang, J.] | - |
dc.contributor.author | Jung, S.J.[Jung, S.J.] | - |
dc.contributor.author | Jeong, S.[Jeong, S.] | - |
dc.contributor.author | Heo, J.[Heo, J.] | - |
dc.contributor.author | Shin, H.[Shin, H.] | - |
dc.contributor.author | Ham, T.J.[Ham, T.J.] | - |
dc.contributor.author | Lee, J.W.[Lee, J.W.] | - |
dc.date.accessioned | 2021-07-28T13:39:45Z | - |
dc.date.available | 2021-07-28T13:39:45Z | - |
dc.date.created | 2021-05-10 | - |
dc.date.issued | 2020 | - |
dc.identifier.issn | 1063-6897 | - |
dc.identifier.uri | https://scholarworks.bwise.kr/skku/handle/2021.sw.skku/6760 | - |
dc.description.abstract | Object serialization and deserialization (S/D) is an essential feature for efficient communication between distributed computing nodes with potentially non-uniform execution environments. S/D operations are widely used in big data analytics frameworks for remote procedure calls and massive data transfers like shuffles. However, frequent S/D operations incur significant performance and energy overheads as they must traverse and process a large object graph. Prior approaches improve S/D throughput by effectively hiding disk or network I/O latency with computation, increasing compression ratio, and/or application-specific customization. However, inherent dependencies in the existing (de)serialization formats and algorithms eventually become the major performance bottleneck. Thus, we propose Cereal, a specialized hardware accelerator for memory object serialization. By co-designing the serialization format with hardware architecture, Cereal effectively utilizes abundant parallelism in the S/D process to deliver high throughput. Cereal also employs an efficient object packing scheme to compress metadata such as object reference offsets and a space-efficient bitmap representation for the object layout. Our evaluation of Cereal using both a cycle-level simulator and synthesizable Chisel RTL demonstrates that Cereal delivers 43.4× higher average S/D throughput than 88 other S/D libraries on Java Serialization Benchmark Suite. For six Spark applications Cereal achieves 7.97× and 4.81× speedups on average for S/D operations over Java built-in serializer and Kryo, respectively, while saving S/D energy by 227.75× and 136.28×. © 2020 IEEE. | - |
dc.language | 영어 | - |
dc.language.iso | en | - |
dc.publisher | Institute of Electrical and Electronics Engineers Inc. | - |
dc.title | A Specialized Architecture for Object Serialization with Applications to Big Data Analytics | - |
dc.type | Article | - |
dc.contributor.affiliatedAuthor | Jang, J.[Jang, J.] | - |
dc.identifier.doi | 10.1109/ISCA45697.2020.00036 | - |
dc.identifier.scopusid | 2-s2.0-85091993304 | - |
dc.identifier.wosid | 000617734800025 | - |
dc.identifier.bibliographicCitation | Proceedings - International Symposium on Computer Architecture, v.2020-May, pp.322 - 334 | - |
dc.relation.isPartOf | Proceedings - International Symposium on Computer Architecture | - |
dc.citation.title | Proceedings - International Symposium on Computer Architecture | - |
dc.citation.volume | 2020-May | - |
dc.citation.startPage | 322 | - |
dc.citation.endPage | 334 | - |
dc.type.rims | ART | - |
dc.type.docType | Proceedings Paper | - |
dc.description.journalClass | 1 | - |
dc.description.isOpenAccess | N | - |
dc.description.journalRegisteredClass | scopus | - |
dc.relation.journalResearchArea | Computer Science | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Hardware & Architecture | - |
dc.relation.journalWebOfScienceCategory | Computer Science, Theory & Methods | - |
dc.subject.keywordPlus | Big data | - |
dc.subject.keywordPlus | Computer hardware | - |
dc.subject.keywordPlus | Computer software | - |
dc.subject.keywordPlus | Data Analytics | - |
dc.subject.keywordPlus | Data transfer | - |
dc.subject.keywordPlus | Java programming language | - |
dc.subject.keywordPlus | Metadata | - |
dc.subject.keywordPlus | Network architecture | - |
dc.subject.keywordPlus | Bitmap representation | - |
dc.subject.keywordPlus | Efficient communications | - |
dc.subject.keywordPlus | Execution environments | - |
dc.subject.keywordPlus | Hardware architecture | - |
dc.subject.keywordPlus | Object serialization | - |
dc.subject.keywordPlus | Performance bottlenecks | - |
dc.subject.keywordPlus | Remote Procedure Call | - |
dc.subject.keywordPlus | Specialized hardware | - |
dc.subject.keywordPlus | Advanced Analytics | - |
dc.subject.keywordAuthor | Apache Spark | - |
dc.subject.keywordAuthor | Data analytics | - |
dc.subject.keywordAuthor | Domain-specific architecture | - |
dc.subject.keywordAuthor | Hardware-software co-design | - |
dc.subject.keywordAuthor | Object serialization | - |
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.