Lecture Notes in Computer Science 5415 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 UniversityofDortmund,Germany MadhuSudan MassachusettsInstituteofTechnology,MA,USA DemetriTerzopoulos UniversityofCalifornia,LosAngeles,CA,USA DougTygar UniversityofCalifornia,Berkeley,CA,USA GerhardWeikum Max-PlanckInstituteofComputerScience,Saarbruecken,Germany Eduardo César Michael Alexander Achim Streit Jesper Larsson Träff Christophe Cérin Andreas Knüpfer Dieter Kranzlmüller Shantenu Jha (Eds.) Euro-Par 2008 Workshops – Parallel Processing VHPC 2008, UNICORE 2008, HPPC 2008, SGS 2008, PROPER 2008, ROIA 2008, and DPA 2008 Las Palmas de Gran Canaria, Spain,August 25-26, 2008 Revised Selected Papers 1 3 VolumeEditors EduardoCésar UniversidadAutónomadeBarcelona,Spain E-mail:[email protected] MichaelAlexander WirtschaftsuniversitätWien,Austria E-mail:[email protected] AchimStreit JülichSupercomputingCentre,Germany E-mail:[email protected] JesperLarssonTräff NECLaboratoriesEurope,SanktAugustin,Germany E-mail:[email protected] ChristopheCérin UniversitédeParisNord,LIPN,France E-mail:[email protected] AndreasKnüpfer TechnischeUniversitätDresden,Germany E-mail:[email protected] DieterKranzlmüller LMUMünchen,Germany E-mail:[email protected] ShantenuJha LouisianaStateUniversity,USA E-mail:[email protected] LibraryofCongressControlNumber:Appliedfor CRSubjectClassification(1998):C.1-4,D.1-4,F.1-3,G.1-2,H.2 LNCSSublibrary:SL1–TheoreticalComputerScienceandGeneralIssues ISSN 0302-9743 ISBN-10 3-642-00954-9SpringerBerlinHeidelbergNewYork ISBN-13 978-3-642-00954-9SpringerBerlinHeidelbergNewYork Thisworkissubjecttocopyright.Allrightsarereserved,whetherthewholeorpartofthematerialis concerned,specificallytherightsoftranslation,reprinting,re-useofillustrations,recitation,broadcasting, reproductiononmicrofilmsorinanyotherway,andstorageindatabanks.Duplicationofthispublication orpartsthereofispermittedonlyundertheprovisionsoftheGermanCopyrightLawofSeptember9,1965, initscurrentversion,andpermissionforusemustalwaysbeobtainedfromSpringer.Violationsareliable toprosecutionundertheGermanCopyrightLaw. springer.com ©Springer-VerlagBerlinHeidelberg2009 PrintedinGermany Typesetting:Camera-readybyauthor,dataconversionbyScientificPublishingServices,Chennai,India Printedonacid-freepaper SPIN:12643462 06/3180 543210 Preface Parallel and distributed processing, although within the focus of computer science research for a long time, is gaining more and more importance in a wide spectrum of applications. These proceedings aim to demonstrate the use of parallel and distributed processing concepts in different application fields, and attempt to spark interest in novel research directions to parallel and high-performance computing research in general. The objective of these workshops is to specifically address researchers coming from university, industry and governmental research organizations and application-oriented companies in order to close the gap between purely scientific research and the applicabil- ity of the research ideas to real-life problems. Euro-Par is an annual series of international conferences dedicated to the promo- tion and advancement of all aspects of parallel and distributed computing. The 2008 event was the 14th issue of the conference. Euro-Par has for a long time been eager to attract colocated events sharing the same goal of promoting the devel- opment of parallel and distributed computing, both as an industrial technique and an academic discipline, extending the frontier of both the state of the art and the state of the practice. Since 2006, Euro-Par has been offering researchers the chance to colo- cate advanced technical workshops back-to-back with the main conference. This is for a mutual benefit: the workshops can take advantage of all technical and social facili- ties that are set up for the conference, so that the organizational tasks are kept to a minimal level; the conference can rely on workshops to experiment with specific areas of research that are not yet mature enough, or too specific, to lead to an official, full-fledged topic at the conference. The 2006 and 2007 events were quite successful, and were extended to a larger size in 2008, where nine events were colocated with the main Euro-Par Conference: • CoreGRID Symposium is the major annual event of the CoreGRID Euro- pean Research Network on Foundations, Software Infrastructures and Appli- cations for large-scale distributed, grid and peer-to-peer technologies. It is also an opportunity for a number of CoreGRID Working Groups to organize their regular meetings. The proceedings have been published in a specific volume of the Springer CoreGRID series, Towards Next Generation Grids. • GECON 2008 is the 5th International Workshop on Grid Economic and Busi- ness Model. Euro-Par was eager to attract an event about this very important aspect of grid computing, which has often been overlooked by scientific re- searchers of the field. Its proceedings are published in a separate volume of Springer’s Lecture Notes in Computer Science series. • VHPC 2008 is the Workshop on Virtualization/Xen in High-Performance Cluster and Grid Computing. Virtual machine monitors (VMMs) are now VI Preface integrated with a variety of operating systems and are moving out of research labs into scientific, educational and operational usage. This workshop aimed to bring together researchers and practitioners active in exploring the application of virtualization in distributed and high-performance cluster and grid computing environments. This was a unique opportunity for the Euro-Par community to make connections with this very active research domain. • UNICORE Summit 2008 brought together researchers and practitioners working with UNICORE in the areas of grid and distributed computing, to exchange and share their experiences, new ideas and latest research results on all aspects of UNICORE. The UNICORE grid technology provides a seamless, secure and intuitive access to distributed grid resources. This was the fourth meeting of the UNICORE community, after a meeting in Sophia-Antipolis, France, in 2005, and a colocated meeting at Euro-Par 2006 in Dresden, Germany, in 2006, and Euro-Par 2007 in Rennes, France. • HPPC 2008 is the Second Workshop on Highly Parallel Processing on a Chip. With a number of both general and special purpose multi-core proces- sors already on the market, it is foreseeable that new designs with a substan- tial number of processing cores will emerge to meet demands for extremely high performance, dependability and controllable power consumption in mo- bile and embedded devices, and in response to the convergence of communi- cation, media and compute devices. The HPPC workshop aims to be(come) a forum for discussion of the major challenges to architecture, language and compiler design, algorithms and application developments, in order to fully (or acceptably) exploit the raw compute power of multi-core processors with a significant amount of parallelism. • SGS 2008 is the First Workshop on Secure, Trusted, Manageable and Control- lable Grid Services. It refers to the notions of security, the way we manage such large systems and the way we control the grid system. For instance, the word 'controllable' means: how we measure the activity of the grid and how we report it. The word 'manageable' means: 'how we deploy the grid architecture, the grid softwares, and how we start jobs (under controllable events such as the availabil- ity of resources). The word 'security' refers to the traditional fields of authentica- tion, fault tolerance but refers also to safe execution (how to certify results, how to adapt computation according to some metric). Moreover, all these services should collaborate making the building of middleware a challenging problem. The building of chains of trust between software components as well as the inte- gration of security and privacy mechanisms across multiple autonomous and/or heterogeneous grid platforms are key challenges for the community. • The PROPER 2008 workshop was organized on behalf of the Virtual Insti- tute for High Productivity Supercomputing (VI-HPS), which aims at improv- ing the quality and accelerating the development process of complex simulation codes in science and engineering that are being designed to run on highly paral- lel computer systems. One part of this mission is the development of integrated state-of-the-art programming tools for high-performance computing that assist Preface VII programmers in diagnosing programming errors and optimizing the perform- ance of their applications. Accordingly, the workshop topics cover tools for parallel program develop- ment and analysis as well as general performance measurement and evaluation approaches. Last but not least, it includes success stories about optimization or parallel scalability achieved using the tools. In particular, the workshop wants to stimulate discussion between tool developers and experts on one hand and tool users and application developers on the other hand. Furthermore, it especially supports younger researchers to present their work. • ROIA 2008 is the First International Workshop on Real-Time Online Inter- active Applications on the Grid. It aimed to bring together researchers from the domain of ROIAs and grid computing in order to exchange knowledge, experiences, ideas and concepts for combining both fields. The event was closely related to the research perfomed in the European edutain@grid project. • DPA 2008 aimed to determine where programming abstractions are important and where non-programmatic abstractions are likely to make greater impact in enabling applications to effectively utilize distributed infrastructure. This work- shop will have a balance of applications and topical infrastructure developments (such as abstractions for Clouds). The reader will find in this volume the proceedings of the last seven events. Hosting Euro-Par 2008 and these colocated events in Las Palmas de Gran Canaria would not have been possible without the support and the help of different institutions and numerous people. Although we are thankful to many more people, we are particularly grateful to the workshop organizers: Martti Forsell and Jesper Larsson Träff for HPPC 2008; Achim Streit and Wolfgang Ziegler for UNICORE Summit 2008; and Michael Alexander and Stephen Childs for VHPC 2008. It has been a pleasure to collaborate with them on this project. We particularly thank them for their interest in our proposal and their trust and availability along the entire preparation process. Euro-Par 2008 was hosted on the university campus and we would like to thank the University Institute for Intelligent Systems and Numerical Applications in Engineering of the Universidad de Las Palmas de Gran Canaria for the support and infrastructure. We gratefully acknowledge the great organizational support of the Computer Architecture and Operating Systems Department of the Universidad Autónoma de Barcelona. We would also like to thank the Cabildo de Gran Canaria and the City Council of Las Palmas de Gran Canaria for they institutional support. Finally, we are grateful to Springer for agreeing to publish the proceedings of these seven workshops in a specific volume of its Lecture Notes in Computer Science series. We are definitely eager to pursue this collaboration. It has been a great pleasure to work together on this project in Las Palmas de Gran Canaria. VIII Preface We hope that the current proceedings are beneficial for the sustainable growth and awareness of parallel and distributed computing concepts in future applications. Dec ember 2008 Eduardo César Michael Alexander Achim Streit Jesper Larsson Träff Christophe Cérin Andreas Knüpfer Dieter Kranzlmüller Shantenu Jha Organization Euro-Par Steering Committee Chair Christian Lengauer University of Passau, Germany Vice-Chair Luc Bougé ENS Cachan, France European Representatives José Cunha New University of Lisbon, Portugal Marco Danelutto University of Pisa, Italy Rainer Feldmann University of Paderborn, Germany Christos Kaklamanis Computer Technology Institute, Greece Anne-Marie Kermarrec IRISA, Rennes, France Paul Kelly Imperial College, UK Harald Kosch University of Klagenfurt, Austria Thomas Ludwig University of Heidelberg, Germany Emilio Luque University Autonoma of Barcelona, Spain Luc Moreau University of Southampton, UK Wolfgang Nagel Dresden University of Technology, Germany Rizos Sakellariou University of Manchester, UK Non-European Representatives Jack Dongarra University of Tennessee at Knoxville, USA Shinji Tomita Kyoto University, Japan Honorary Members Ron Perrott Queen's University Belfast, UK Karl Dieter Reinartz University of Erlangen-Nuremberg, Germany Observers Domingo Benitez University of Las Palmas, Gran Canaria, Spain Henk Sips Delft University of Technology, The Netherlands X Organization Euro-Par 2008 Local Organization Conference Co-chairs Emilio Luque UAB General Chair Domingo Benítez ULPGC Vice-Chair Tomàs Margalef UAB Vice-Chair Local Organizing Committee Eduardo César (UAB) Ana Cortés (UAB) Daniel Franco (UAB) Elisa Heymann (UAB) Anna Morajko (UAB) Juan Carlos Moure (UAB) Dolores Rexachs (UAB) Miquel Àngel Senar (UAB) Joan Sorribes (UAB) Remo Suppi (UAB) Web and Technical Support Daniel Ruiz (UAB) Javier Navarro (UAB) Euro-Par 2008 Workshop Program Committees Third Workshop on Virtualization in High-Performance Cluster and Grid Computing (VHPC 2008) Program Chairs Michael Alexander (Chair) WU Vienna, Austria Stephen Childs (Co-chair) Trinity College, Dublin, Ireland Program Committee Jussara Almeida Federal University of Minas Gerais, Brazil Padmashree Apparao Intel Corp., USA Hassan Barada Etisalat University College, UAE Volker Buege University of Karlsruhe, Germany Simon Crosby Xensource, UK Marcus Hardt Forschungszentrum Karlsruhe, Germany Sverre Jarp CERN, Switzerland Krishna Kant Intel Corporation, USA Yves Kemp University of Karlsruhe, Germany Organization XI Naoya Maruyama Tokyo Institute of Technology, Japan Jean-Marc Menaud Ecole des Mines de Nantes, France José E. Moreira IBM T.J. Watson Research Center, USA Jose Renato Santos HP Labs, USA Andreas Schabus Microsoft , Austria Yoshio Turner HP Labs, USA Andreas Unterkircher CERN, Switzerland Dongyan Xu, Purdue University, USA UNICORE Summit 2008 Program Chairs Achim Streit Forschungszentrum Jülich, Germany Wolfgang Ziegler Fraunhofer Gesellschaft SCAI, Germany Program Committee Agnes Ansari CNRS-IDRIS, France Rosa Badia Barcelona Supercomputing Center, Spain Thomas Fahringer University of Innsbruck, Austria Donal Fellows University of Manchester, UK Anton Frank LRZ Munich, Germany Edgar Gabriel University of Houston, USA Alfred Geiger T-Systems, Germany Fredrik Hedman KTH-PDC, Sweden Odej Kao Technical University of Berlin, Germany Paolo Malfetti CINECA, Italy Ralf Ratering Intel GmbH, Germany Mathilde Romberg Forschungszentrum Jülich, Germany Bernd Schuller Forschungszentrum Jülich, Germany David Snelling Fujitsu Laboratories of Europe, UK Thomas Soddemann Max-Planck-Institut für Plasmaphysik - RZG, Germany Stefan Wesner University of Stuttgart - HLRS, Germany Ramin Yahyapour University of Dortmund, Germany Additional Reviewers Max Berger Kassian Plankensteiner