Please use this identifier to cite or link to this item:
Title: Performance evaluation of group communication architectures in large scale systems using MPI
Authors: Erciyeş, Kayhan
Dağdeviren, Orhan
Payli, Reşat Ümit
Keywords: Large scale systems
Fault tolerant computer systems
Pipeline processing systems
User interfaces
Issue Date: 2006
Publisher: Springer Verlag
Source: Erciyeş, K., Dağdeviren, O., and Payli, R. Ü. (2006). Performance evaluation of group communication architectures in large scale systems using MPI. Lecture Notes in Computer Science, 4276 LNCS(PART 2), 1422-1432. doi:10.1007/11914952_28
Abstract: Group communication is an important paradigm for fault tolerance in large scale systems. We describe various group architectures as pipelined, hierarchical, daisy and hypercube groups each consisting of separate clusters, investigate the theoretical performance bounds of these architectures and evaluate their experimental performances using MPI group communication primitives. We first derive time bounds for multicast message deliveries in these architectures and then provide tests to measure the times taken for the same operation. The multicast message delivery times are tested against the number of clusters within a group and the size of the multicast message. We conclude that daisy architecture is favorable both in terms of delivery times and message sizes theoretically and experimentally.
Description: OTM Confederated International Conferences, CoopIS, DOA, GADA, and ODBASE 2006; Montpellier; France; 29 October 2006 through 3 November 2006
ISBN: 9783540482741
ISSN: 0302-9743
Appears in Collections:Computer Engineering / Bilgisayar Mühendisliği
Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collection
WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collection

Files in This Item:
File Description SizeFormat 
2248.pdfConference Paper526.11 kBAdobe PDFThumbnail
Show full item record

CORE Recommender


checked on Dec 3, 2022

Page view(s)

checked on Nov 28, 2022


checked on Nov 28, 2022

Google ScholarTM



Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.