• Türkçe
    • English
  • English 
    • Türkçe
    • English
  • Login
View Item 
  •   DSpace@MEF
  • Fakülteler
  • Mühendislik Fakültesi
  • Bilgisayar Mühendisliği | Computer Engineering
  • MF, BM, Makale Koleksiyonu
  • View Item
  •   DSpace@MEF
  • Fakülteler
  • Mühendislik Fakültesi
  • Bilgisayar Mühendisliği | Computer Engineering
  • MF, BM, Makale Koleksiyonu
  • View Item
JavaScript is disabled for your browser. Some features of this site may not work without it.
Advanced Search

Array BP-XOR codes for hierarchically distributed matrix multiplication

Thumbnail

View/Open

Full Text - Article (765.2Kb)

Access

info:eu-repo/semantics/closedAccess

Date

2021

Author

Arslan, Şuayb Şefik

Metadata

Show full item record

Citation

Arslan, S. S. (02 December 2021). Array BP-XOR Codes for Hierarchically Distributed Matrix Multiplication. IEEE Transactions on Information Theory, pp. 1–17. https://doi.org/10.1109/tit.2021.3132043 ‌ ‌

Abstract

A novel fault-tolerant computation technique based on array Belief Propagation (BP)-decodable XOR (BP-XOR) codes is proposed for distributed matrix-matrix multiplication. The proposed scheme is shown to be configurable and suited for modern hierarchical compute architectures such as Graphical Processing Units (GPUs) equipped with multiple nodes, whereby each has many small independent processing units with increased core-to-core communications. The proposed scheme is shown to outperform a few of the well–known earlier strategies in terms of total end-to-end execution time while in presence of slow nodes, called stragglers. This performance advantage is due to the careful design of array codes which distributes the encoding operation over the cluster (slave) nodes at the expense of increased master-slave communication. An interesting trade-off between end-to-end latency and total communication cost is precisely described. In addition, to be able to address an identified problem of scaling stragglers, an asymptotic version of array BP-XOR codes based on projection geometry is proposed at the expense of some computation overhead. A thorough latency analysis is conducted for all schemes to demonstrate that the proposed scheme achieves order-optimal computation in both the sublinear as well as the linear regimes in the size of the computed product from an end-to-end delay perspective.

URI

https://doi.org/10.1109/tit.2021.3132043
https://hdl.handle.net/20.500.11779/1597

Collections

  • Araştırma Çıktıları, Scopus İndeksli Yayınlar Koleksiyonu [455]
  • Araştırma Çıktıları, WOS İndeksli Yayınlar Koleksiyonu [482]
  • MF, BM, Makale Koleksiyonu [27]



DSpace software copyright © 2002-2015  DuraSpace
Contact Us | Send Feedback
Theme by 
@mire NV
 

 




| Instruction | Guide | Contact |

DSpace@MEF

by OpenAIRE

sherpa/romeo

Browse

All of DSpaceCommunities & CollectionsBy Issue DateAuthorsInstitution AuthorTitlesORCIDSubjectsTypeLanguageDepartmentCategoryPublisherAccess TypeThis CollectionBy Issue DateAuthorsInstitution AuthorTitlesORCIDSubjectsTypeLanguageDepartmentCategoryPublisherAccess Type

My Account

LoginRegister

Statistics

View Google Analytics Statistics

DSpace software copyright © 2002-2015  DuraSpace
Contact Us | Send Feedback
Theme by 
@mire NV
 

 


|| Guide|| Instruction || Library || MEF University || OAI-PMH ||

MEF University Library, İstanbul, Turkey
If you find any errors in content please report us

Creative Commons License
MEF University Institutional Repository is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 4.0 Unported License..

DSpace@MEF:


DSpace 6.2

tarafından İdeal DSpace hizmetleri çerçevesinde özelleştirilerek kurulmuştur.