Distributed and Parallel Computing: 6th International Conference on Algorithms and Architectures for Parallel Processing, ICA3PP, Melbourne, Australia, October 2-3, 2005. Proceedings

462 Pages·2005·8.23 MB·English

by  Fengjung Liu

Lecture Notes in Computer Science 3719
CommencedPublicationin1973
FoundingandFormerSeriesEditors:
GerhardGoos,JurisHartmanis,andJanvanLeeuwen
EditorialBoard
DavidHutchison
LancasterUniversity,UK
TakeoKanade
CarnegieMellonUniversity,Pittsburgh,PA,USA
JosefKittler
UniversityofSurrey,Guildford,UK
JonM.Kleinberg
CornellUniversity,Ithaca,NY,USA
FriedemannMattern
ETHZurich,Switzerland
JohnC.Mitchell
StanfordUniversity,CA,USA
MoniNaor
WeizmannInstituteofScience,Rehovot,Israel
OscarNierstrasz
UniversityofBern,Switzerland
C.PanduRangan
IndianInstituteofTechnology,Madras,India
BernhardSteffen
UniversityofDortmund,Germany
MadhuSudan
MassachusettsInstituteofTechnology,MA,USA
DemetriTerzopoulos
NewYorkUniversity,NY,USA
DougTygar
UniversityofCalifornia,Berkeley,CA,USA
MosheY.Vardi
RiceUniversity,Houston,TX,USA
GerhardWeikum
Max-PlanckInstituteofComputerScience,Saarbruecken,Germany
Michael Hobbs Andrzej M. Goscinski
Wanlei Zhou (Eds.)
Distributed
and
Parallel Computing
6th International Conference on Algorithms and
Architectures for Parallel Processing, ICA3PP
Melbourne, Australia, October 2-3, 2005
Proceedings
1 3
VolumeEditors
MichaelHobbs
AndrzejM.Goscinski
WanleiZhou
DeakinUniversity
SchoolofInformationTechnology
Geelong,Victoria3217
Australia
E-mail:{mick,ang,wanlei}@deakin.edu.au
LibraryofCongressControlNumber:2005933043
CRSubjectClassification(1998):D,F.1-3.C,I.6
ISSN 0302-9743
ISBN-10 3-540-29235-7SpringerBerlinHeidelbergNewYork
ISBN-13 978-3-540-29235-7SpringerBerlinHeidelbergNewYork
Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis
concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting,
reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication
orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965,
initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable
toprosecutionundertheGermanCopyrightLaw.
SpringerisapartofSpringerScience+BusinessMedia
springeronline.com
©Springer-VerlagBerlinHeidelberg2005
PrintedinGermany
Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India
Printedonacid-freepaper SPIN:11564621 06/3142 543210
Preface
There are many applications that require parallel and distributed processing to allow
complicated engineering, business and research problems to be solved in a reasonable
time. Parallel and distributed processing is able to improve company profit, lower
costs of design, production, and deployment of new technologies, and create better
business environments. The major lesson learned by car and aircraft engineers, drug
manufacturers, genome researchers and other specialist is that a computer system is a
very powerful tool that is able to help them solving even more complicated problems.
That has led computing specialists to new computer system architecture and
exploiting parallel computers, clusters of clusters, and distributed systems in the form
of grids. There are also institutions that do not have so complicated problems but
would like to improve profit, lower costs of design and production by using parallel
and distributed processing on clusters.
In general to achieve these goals, parallel and distributed processing must become
the computing mainstream. This implies a need for new architectures of parallel and
distributed systems, new system management facilities, and new application
algorithms. This also implies a need for better understanding of grids and clusters,
and in particular their operating systems, scheduling algorithms, load balancing,
heterogeneity, transparency, application deployment, which is of the most critical
importance for their development and taking them by industry and business.
ICA3PP has been a premier conference that has brought together researchers and
practitioners from academia, industry and governments around the world to advance
the theories and technologies of parallel and distributed computing. Previously,
ICA3PP conferences have been held successfully in Brisbane, Singapore, Melbourne,
Hong Kong and Beijing.
ICA3PP 2005 returned to Melbourne with the main focus on the most critical areas
of parallel and distributed computing: operating systems and middleware, fault-
tolerant systems, scheduling and load balancing, algorithms, tools and environments,
and communication and networks.
In total, the conference received 98 papers from researchers and practitioners from
15 countries. Each paper was reviewed by at least three internationally renowned
referees and selected based on their originality, significance, correctness, relevance,
and clarity of presentation. Among the high quality submissions, 27 long papers and
25 short papers were accepted. All of the selected papers are included in the
proceedings. After the conference, the proceedings editors will recommend some
high quality papers from the conference to be published in a special issue of an
international journal.
We are delighted to be able to host two well-known international scholars,
Professor Ian Foster and Professor Zhiwei Xu, who delivered the keynote speeches.
VI Preface
We would like to take this opportunity to thank all the authors for their
submissions to the conference. Many of them have traveled some distance to
participate in the conference. We also thank the Program Committee members and
additional reviewers for their efforts in reviewing the large number of papers. Thanks
also go to the local conference organizers for their great support.
Last but not least, we would like to express our gratitude to all of the organizations
who have supported our efforts to bring the conference to fruition. We are grateful to
the IEEE Technical Committee on Scalable Computing for the cooperation; and to
Deakin University, NICTA and Alexander Technology for their sponsorships and
assistance.
October 2005 Michael Hobbs, Andrzej Goscinski and Wanlei Zhou
Melbourne,
Organization
This conference was organized by the School of IT, Deakin University, Australia; and
Martin Lack & Associates, Australia. Sponsorship was provided by the School of IT,
Deakin University, Australia; the Faculty of Science and Technology, Deakin
University, Australia; National ICT Australia (NICTA); and Alexander Technology,
Australia.
Conference Chairs
Andrzej M. Goscinski (Deakin University, Australia)
Wanlei Zhou (Deakin University, Australia)
Program Chair
Michael J. Hobbs (Deakin University, Australia)
Program Committee
Jemal Abawajy (Deakin University, Australia)
David Abromson (Monash University, Australia)
Akkihebbal Ananda (National University of Singapore, Singapore)
Bill Appelbe (RMIT University, Australia)
Mark Baker (Portsmouth University, UK)
Amnon Barak (Hebrew University of Jerusalem, Israel)
Arndt Bode (Technical University of Munich, Germany)
Peter Brezany (University of Vienna, Austria)
Marian Bubak (AGH University of Science and Technology., Cracow, Poland)
Raj Buyya (University of Melbourne, Australia)
Jianning Cao (Hong Kong Polytechnic University, Hong Kong)
Samuel Chanson (Hong Kong University of Science and Technology, Hong Kong)
Jianer Chen (Texas A&M University, USA)
Francis Chin (University of Hong Kong, Hong Kong)
Toni Cortes (University of Politecnica de Catalunya, Spain)
Brian d'Auriol (University of Texas at El Paso, USA)
Xiaote Deng (City University of Hong Kong, Hong Kong)
Robert Dew (Deakin University, Australia)
Jack Dongarra (University of Tennessee, USA)
Ding-Zhu Du (University of Minnesota, USA)
Wen Gao (Inst. of Computing Technology, China)
Al Giest (Oak Ridge Nation Labs, USA)
Minyi Guo (University of Aizu, Japan)
VIII Organization
Salim Hariri (Syracuse University, USA)
Louis Hertzberger (University of Amsterdam, The Netherland)
Shi-Jinn Horng (National Taiwan University of Science and Technology, Taiwan)
Ali Hurson (Pennsylvania State University, USA)
Weijia Jia (City University of Hong Kong, Hong Kong)
Xiaohua Jia (City University of Hong Kong, Hong Kong)
Hai Jin (Huazhong University of Science and Technology, China)
Peter Kacsuk (MTA SZTAKI Research Inst., Hungry)
Krishna Kavi (The University of North Texas, USA)
Zvi Kedem (New York University, USA)
Wayne Kelly (Queensland University of Tech., Australia)
Tohru Kikuno (Osaka University, Japan)
Jacek Kitowski (University of Mining and Metallurgy, Poland)
Domenico Laforenza (ISTI-CNR, Italy)
Laurent Leferve (INRIA, Lyon, France)
Keqin Li (State University of New York at New Paltz, USA)
Zhi Yong Liu (National Natural Science Foundation, China)
Thomas Ludwig (University of Heidelberg, Germany)
George Mohay (Queensland University of Tech., Australia)
Christine Morin (IRISA/INRIA, France)
Edgar Nett (Otto-von-Guericke University, Germany)
Yi Pan (Georgia State University USA)
Marcin Paprzycki (Oklahoma State University)
Sushil K. Prasad (Georgia State University USA)
Rajeev Raje (Purdue University, USA)
Michel Raynal (IRISA, France)
Justin Rough (Deakin University, Australia)
Srinivas Sampalli (Dalhousie University, Canada)
Edwin Sha (University of Texas at Dallas, USA)
Behrooz Shirazi (University of Texas Arlington, USA)
Jackie Silcock (Deakin University, Australia)
Beth Simon (University of San Diego, USA)
Chengzheng Sun (Griffith University, Australia)
Jiachang Sun (Institute of Software, China)
Vaidy Sunderam (Emory University, Atlanta, USA)
Yong-Meng Teo (National University of Singapore, Singapore)
Alistair Veitch (Hewlett Packard Labs, California, USA)
Greg Wickham (GrangeNet, Australia)
Jie Wu (Florida Atlantic University, USA)
Yue Wu (University of Electronic Science and Technology, China)
Roman Wyrzykowski (Czestochowa University of Technology, Poland)
Jingling Xue (University of New South Wales, Australia)
Zhiwei Xu (Inst. of Computing Technology, China)
Laurence T. Yang (St. Francis Xavier University, Canada)
Chung-Kwong Yuen (National University of Singapore, Singapore)
Si Q Zheng (University of Texas Dallas, USA)
Weimin Zheng (Tsinghua University, China)
Wei Zhao (Texas A&M University, USA)
Jun Zou (Chinese University of Hong Kong, China)
Albert Zoymaya (University of Sydney, Australia)
Table of Contents
Improving Concurrent Write Scheme in File Server Group
Fengjung Liu, Chu-sing Yang................................... 1
A Comparative Performance Study of Distributed Mutual Exclusion
Algorithms with a Class of Extended Petri Nets
Alexander Kostin, Ljudmila Ilushechkina, Erhan Basri............. 11
A Practical Comparison of Cluster Operating Systems Implementing
Sequential and TransactionalConsistency
StefanFrenz,RenaudLottiaux,Michael Schoettner,ChristineMorin,
Ralph Goeckelmann, Peter Schulthess............................ 23
Clock Synchronization State Graphs Based on Clock Precision Difference
Ying Zhao, Wanlei Zhou, Yingying Zhang, E.J. Lanham,
Jiumei Huang ................................................ 34
A Recursive-Adjustment Co-allocation Scheme in Data Grid
Environments
Chao-Tung Yang, I-Hsien Yang, Kuan-Ching Li,
Ching-Hsien Hsu ............................................. 40
Reducing the Bandwidth Requirements of P2P Keyword Indexing
John Casey, Wanlei Zhou...................................... 50
A Deadline and Budget Constrained Scheduling Algorithm for eScience
Applications on Data Grids
Srikumar Venugopal, Rajkumar Buyya ........................... 60
A Survivability Model for Cluster System
Khin Mi Mi Aung, Kiejin Park, Jong Sou Park................... 73
Localization Techniques for Cluster-Based Data Grid
Ching-Hsien Hsu, Guan-Hao Lin, Kuan-Ching Li,
Chao-Tung Yang.............................................. 83
GridFTP and ParallelTCP Support in NaradaBrokering
Sang Boem Lim, Geoffrey Fox, Ali Kaplan, Shrideep Pallickara,
Marlon Pierce ................................................ 93
2-LayeredMetadata Service Model in Grid Environment
Muzhou Xiong, Hai Jin, Song Wu .............................. 103
X Table of Contents
pKSS: An Efficient Keyword Search System in DHT Peer-to-Peer
Network
Yin Li, Fanyuan Ma, Liang Zhang .............................. 112
A Comparative Study at the Logical Level of Centralised and
Distributed Recovery in Clusters
Andrew Maloney, Andrzej Goscinski ............................ 118
Toward Self Discovery for an Autonomic Cluster
Eric Dines, Andrzej Goscinski.................................. 125
Mining Traces of Large Scale Systems
Christophe C´erin, Michel Koskas ............................... 132
Setup Algorithm of Web Service Composition
YanPing Yang, QingPing Tan, Yong Xiao........................ 139
Self Healing and Self Configuration in a WSRF Grid Environment
Michael Messig, Andrzej Goscinski.............................. 149
Study on Life Cycle Model of Dynamic Composed Web Services
Chen Yanping, Li Zengzhi, Jin Qinxue, Wang Chuang............. 159
Fault-Tolerant Dynamic Job Scheduling Policy
J.H. Abawajy................................................. 165
An Efficient Dynamic Load-Balancing Algorithm in a Large-Scale
Cluster
Bao-Yin Zhang, Ze-Yao Mo, Guang-Wen Yang,
Wei-Min Zheng............................................... 174
Job Scheduling Policy for High Throughput Grid Computing
J.H. Abawajy................................................. 184
High Performance Task Scheduling Algorithm for Heterogeneous
Computing System
E. Ilavarasan, P. Thambidurai, R. Mahilmannan ................. 193
ExecutionEnvironmentsandBenchmarksfortheStudyofApplications’
Scheduling on Clusters
Adam K.L. Wong, Andrzej M. Goscinski ........................ 204
Data Distribution Strategies for Domain Decomposition Applications
in Grid Environments
Beatriz Otero, Jos´e M. Cela, Rosa M. Badia, Jesu´s Labarta ........ 214
Table of Contents XI
Inter-round Scheduling for Divisible Workload Applications
DongWoo Lee, R.S. Ramakrishna ............................... 225
Scheduling Divisible Workloads Using the Adaptive Time Factoring
Algorithm
Tiago Ferreto, C´esar De Rose .................................. 232
Adaptive Policy Triggering for Load Balancing
Dan Feng, Lingfang Zeng ...................................... 240
Parallel Algorithms for Fault-Tolerant Mobile Agent Execution
Jin Yang, Jiannong Cao, Weigang Wu, Cheng-Zhong Xu .......... 246
Design and Multithreading Implementation of the Wave-Front
Algorithm for Constructing Voronoi Diagrams
Grace J. Hwang, Joseph M. Arul, Eric Lin, Chung-Yun Hung ...... 257
A Proposal of Parallel Strategy for Global Wavelet-Based Registration
of Remote-Sensing Images
Haifang Zhou, Yu Tang, Xuejun Yang, Hengzhu Liu............... 267
Performance Analysis of a Parallel Sort Merge Join on Cluster
Architectures
Erich Schikuta................................................ 277
Parallel Clustering on the Star Graph
M. Fazeli, H. Sarbazi-Azad, R. Farivar .......................... 287
Hierarchical ParallelSimulated Annealing and Its Applications
Shiming Xu, Wenguang Chen, Weimin Zheng, Tao Wang,
Yimin Zhang ................................................. 293
Multi-color Difference Schemes of Helmholtz Equation and Its Parallel
Fast Solver over 3-D Dodecahedron Partitions
Jiachang Sun................................................. 301
GridMD: ProgramArchitecture for Distributed Molecular Simulation
Ilya Valuev................................................... 309
Visuel: A Novel Performance Monitoring and Analysis Toolkit for
Cluster and Grid Environments
Kuan-Ching Li, Hsiang-Yao Cheng, Chao-Tung Yang,
Ching-Hsien Hsu, Hsiao-Hsi Wang, Chia-Wen Hsu,
Sheng-Shiang Hung, Chia-Fu Chang, Chun-Chieh Liu,
Yu-Hwa Pan ................................................. 315