Data Management in Grid and Peer-to-Peer Systems: Third International Conference, Globe 2010, Bilbao, Spain, September 1-2, 2010. Proceedings PDF
Preview Data Management in Grid and Peer-to-Peer Systems: Third International Conference, Globe 2010, Bilbao, Spain, September 1-2, 2010. Proceedings
Lecture Notes in Computer Science 6265 CommencedPublicationin1973 FoundingandFormerSeriesEditors: GerhardGoos,JurisHartmanis,andJanvanLeeuwen EditorialBoard DavidHutchison LancasterUniversity,UK TakeoKanade CarnegieMellonUniversity,Pittsburgh,PA,USA JosefKittler UniversityofSurrey,Guildford,UK JonM.Kleinberg CornellUniversity,Ithaca,NY,USA AlfredKobsa UniversityofCalifornia,Irvine,CA,USA FriedemannMattern ETHZurich,Switzerland JohnC.Mitchell StanfordUniversity,CA,USA MoniNaor WeizmannInstituteofScience,Rehovot,Israel OscarNierstrasz UniversityofBern,Switzerland C.PanduRangan IndianInstituteofTechnology,Madras,India BernhardSteffen TUDortmundUniversity,Germany MadhuSudan MicrosoftResearch,Cambridge,MA,USA DemetriTerzopoulos UniversityofCalifornia,LosAngeles,CA,USA DougTygar UniversityofCalifornia,Berkeley,CA,USA GerhardWeikum MaxPlanckInstituteforInformatics,Saarbruecken,Germany Abdelkader Hameurlain Franck Morvan A Min Tjoa (Eds.) Data Management in Grid and Peer-to-Peer Systems Third International Conference, Globe 2010 Bilbao, Spain, September 1-2, 2010 Proceedings 1 3 VolumeEditors AbdelkaderHameurlain FranckMorvan PaulSabatierUniversity InstitutdeRechercheenInformatiquedeToulouse(IRIT) 118,routedeNarbonne,31062ToulouseCedex,France E-mail:{hameur,morvan}@irit.fr AMinTjoa ViennaUniversityofTechnology InstituteofSoftwareTechnologyandInteractiveSystems Favoritenstr.9/188,1040Wien,Austria E-mail:[email protected] LibraryofCongressControlNumber:2010932195 CRSubjectClassification(1998):H.2,C.2,H.4,H.3,D.2,H.5 LNCSSublibrary:SL3–InformationSystemsandApplication,incl.Internet/Web andHCI ISSN 0302-9743 ISBN-10 3-642-15107-8SpringerBerlinHeidelbergNewYork ISBN-13 978-3-642-15107-1SpringerBerlinHeidelbergNewYork Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting, reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965, initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable toprosecutionundertheGermanCopyrightLaw. springer.com ©Springer-VerlagBerlinHeidelberg2010 PrintedinGermany Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India Printedonacid-freepaper 06/3180 Preface Since 2008, Globe has been an annual international conference on data management in grid and peer-to-peer systems. Initially, grid and peer-to-peer systems experienced significant success in scientific and file sharing applications. Today, these systems cover the management of large, distributed and heterogeneous data. These systems are characterized by high heterogeneity, high autonomy and dynamics of nodes, decen- tralization of control and large-scale distribution of resources. Research on data man- agement in grid and peer-to-peer, relatively recent, aims to scale distributed systems and applications that require effective management of voluminous, large-scale distrib- uted and heterogeneous data. The third edition of the international conference Globe was held in Bilbao, Spain during September 1-2, 2010. Globe provided opportunities for academia or industry researchers to present and discuss the latest research and applications on data man- agement in grid and peer-to-peer systems. Globe 2010 received 26 papers from 15 countries. The reviewing process led to the acceptance of 13 papers for presentation at the conference and inclusion in this LNCS volume. Each paper was reviewed by at least two Program Committee members. The conference would not have been possible without the support of the Program Committee members, external reviewers, Organizing Committee, members of the DEXA conference and the authors. In particular, we would like to thank Gabriela Wagner and Roland Wagner (FAW, University of Linz) for their help in the realiza- tion of this conference. June 2010 Abdelkader Hameurlain Franck Morvan A. Min Tjoa Organization Conference Program Chair Abdelkader Hameurlain IRIT, Paul Sabatier University, Toulouse, France Franck Morvan IRIT, Paul Sabatier University, Toulouse, France A Min Tjoa IFS, Vienna University of Technology, Austria Program Committee Philippe Balbiani IRIT, Paul Sabatier University, Toulouse, France Djamal Benslimane LIRIS, University of Lyon, France Leopoldo Bertossi Carleton University School of Computer Science, Ottawa, Canada Lionel Brunie LIRIS, INSA of Lyon, France Qiming Chen HP Labs, Palo Alto, California, USA Frédéric Cuppens Telecom, Bretagne, France Bruno Defude Telecom INT, Evry, France Kayhan Erciyes Ege University, Izmir, Turkey Shahram Ghandeharizadeh University of Southern California, USA Tasos Gounaris Aristotle University of Thessaloniki, Greece Sergio Ilarri University of Zaragoza, Spain Ismail Khalil Johannes Kepler University, Linz, Austria Gildas Menier LORIA, University of South Bretagne, France Anirban Mondal University of Delhi, India Riad Mokadem IRIT, Paul Sabatier University, Toulouse, France Kjetil Nørvåg Norwegian University of Science and Technology, Trondheim, Norway Claudia Roncancio LIG, Grenoble University, France Florence Sedes IRIT, Paul Sabatier University, Toulouse, France Fabricio A.B. Silva University of Lisbon, Portugal Mário J.G. Silva University of Lisbon, Portugal Hela Skaf, LORIA INRIA Nancy-Grand Est, Nancy University, France David Taniar Monash University, Australia Roland Wagner FAW, University of Linz, Austria Wolfram Wöß FAW, University of Linz, Austria VIII Organization External Reviewers Marek Ciglan Norwegian University of Science and Technology, Trondheim, Norway Christos Doulkeridis Norwegian University of Science and Technology, Trondheim, Norway João B. Rocha-Junior Norwegian University of Science and Technology, Trondheim, Norway Efthymia Tsamoura Aristotle University of Thessaloniki, Greece Table of Contents Data Storage and Management High Throughput Data-Compressionfor Cloud Storage ............... 1 Bogdan Nicolae Merging File Systems and Data Bases to Fit the Grid ................ 13 Yves Denneulin, Cyril Labb´e, Laurent d’Orazio, and Claudia Roncancio Multidimensional Arrays for Warehousing Data on Clouds ............ 26 Laurent d’Orazio and Sandro Bimonte Replication, Caching and Security Continuous Timestamping for Efficient Replication Management in DHTs .......................................................... 38 Reza Akbarinia, Mounir Tlili, Esther Pacitti, Patrick Valduriez, and Alexandre A.B. Lima A Catalog-BasedCaching Strategy for Structured P2P Systems........ 50 Mile Arnedo, Maria del Pilar Villamil, Ricardo Villanueva, Harold Castro, and Laurent d’Orazio Secure Routing Strategies in DHT-Based Systems.................... 62 Ricardo Villanueva, Maria del Pilar Villamil, and Mile Arnedo Data Placement and Performance Evaluation Data Life Time for Different Placement Policies in P2P Storage Systems ........................................................ 75 St´ephane Caron, Fr´ed´eric Giroire, Dorian Mazauric, Julian Monteiro, and St´ephane P´erennes Performance Evaluation of WebRowSet Implementations.............. 89 Alexander Wo¨hrer, Thomas Lustig, and Peter Brezany P2PIRB: Benchmarking Framework for P2PIR ...................... 100 Saloua Zammali and Khedija Arour Verification of General and Cyclic Covering Arrays Using Grid Computing...................................................... 112 Himer Avila-George, Jose Torres-Jimenez, Vicente Herna´ndez, and Nelson Rangel-Valdez X Table of Contents Data Stream Systems and Large Scale Distributed Applications Scale Out Paralleland Distributed CDR Stream Analytics ............ 124 Qiming Chen and Meichun Hsu A Wide Area Network Optimization Service for Effective Access to Heterogeneous DataGrid Infrastructures ............................ 137 Francesco Palmieri and Silvio Pardi A Profile-Based Aggregation Model in a Peer-To-Peer Information Retrieval System................................................. 148 Rim Mghirbi, Khedija Arour, Yahya Slimani, and Bruno Defude Author Index.................................................. 161 High Throughput Data-Compression for Cloud Storage Bogdan Nicolae University of Rennes1 IRISA,Rennes, France Abstract. As data volumes processed by large-scale distributed data- intensive applications grow at high-speed, an increasing I/O pressure is put on the underlying storage service, which is responsible for data management. One particularly difficult challenge, that the storage ser- vice has to deal with, is to sustain a high I/O throughput in spite of heavy access concurrency to massive data. In order to do so, massively parallel data transfers need to be performed, which invariably lead to a high bandwidth utilization. With the emergence of cloud computing, dataintensiveapplicationsbecomeattractiveforawidepublicthatdoes not have the resources to maintain expensive large scale distributed in- frastructures to run such applications. In this context, minimizing the storage space and bandwidth utilization is highly relevant, as these re- sourcesarepaidforaccordingtotheconsumption.Thispaperevaluates the trade-off resulting from transparently applying data compression to conservestoragespaceandbandwidthatthecostofslightcomputational overhead. We aim at reducing the storage space and bandwidth needs with minimal impact on I/O throughput when under heavy access con- currency. Our solution builds on BlobSeer, a highly parallel distributed datamanagementservicespecificallydesignedtoenablereading,writing andappendinghugedatasequencesthatarefragmentedanddistributed at a large scale. We demonstrate the benefits of our approach by per- forming extensiveexperimentations on theGrid’5000 testbed. 1 Introduction Astherate,scaleandvarietyofdataincreasesincomplexity,theneedforflexible applications that can crunch huge amounts of heterogeneous data (such as web pages, online transaction records, access logs, etc.) fast and cost-effective is of utmost importance. Such applications are data-intensive: in a typical scenario, they continuously acquiremassivedatasets(e.g.bycrawlingtheweboranalyzingaccesslogs)while performing computations over these changing datasets (e.g. building up-to-date search indexes). In order to achieve scalability and performance, data acquisi- tions and computations need to be distributed at large scale in infrastructures comprising hundreds and thousands of machines [1]. However, such large scale infrastructure are expensive and difficult to main- tain. The emerging cloud computing model [20,2] is gaining serious interest A.Hameurlain,F.Morvan,andA.MinTjoa(Eds.): Globe2010,LNCS6265,pp. 1–12,2010. (cid:2)c Springer-VerlagBerlinHeidelberg2010
The list of books you might like
