Parallel Computing Technologies

The PaCT-2003 (Parallel Computing Technologies) conference was a four-day conference held in Nizhni Novgorod on September 15–19, 2003. This was the 7th international conference of the PaCT series, organized in Russia every odd year. The ?rst conference, PaCT-91, was held in Novosibirsk (Academgorodok), September 7–11, 1991. The next PaCT conferences were held in: Obninsk (near Moscow),30August–4September,1993;St.Petersburg,September12–15,1995; Yaroslavl, September 9–12, 1997; Pushkin (near St. Petersburg) September 6– 10, 1999; and Akademgorodok (Novosibirsk), September 3–7, 2001. The PaCT proceedings are published by Springer-Verlag in the LNCS series. PaCT-2003 was jointly organized by the Institute of Computational - thematics and Mathematical Geophysics of the Russian Academy of Sciences (Novosibirsk) and the State University of Nizhni Novgorod. The purpose of the conference was to bring together scientists working with theory, architectures, software, hardware and solutions of large-scale problems in order to provide integrated discussions on Parallel Computing Technologies. Theconferenceattractedabout100participantsfromaroundtheworld.Authors from 23 countries submitted 78 papers. Of those submitted, 38 papers were selected for the conference as regular ones; there were also 4 invited papers. In addition,anumberofposterswerepresented.Allthepaperswereinternationally reviewed by at least three referees. As usual a demo session was organized for the participants. Many thanks to our sponsors: the Russian Academy of Sciences, the Russian Fund for Basic Research, the Russian State Committee of Higher Education, IBM and Intel (Intel laboratory in Nizhni Novgorod) for their ?nancial support. The organizers highly appreciate the help of the Association Antenne-Provence (France).

Theory.- Mapping Affine Loop Nests: Solving of the Alignment and Scheduling Problems.- Situated Cellular Agents in Non-uniform Spaces.- Accuracy and Stability of Spatial Dynamics Simulation by Cellular Automata Evolution.- Resource Similarities in Petri Net Models of Distributed Systems.- Authentication Primitives for Protocol Specifications.- An Extensible Coloured Petri Net Model of a Transport Protocol for Packet Switched Networks.- Parallel Computing for Globally Optimal Decision Making.- Parallelization of Alternating Direction Implicit Methods for Three-Dimensional Domains.- Interval Approach to Parallel Timed Systems Verification.- An Approach to Assessment of Heterogeneous Parallel Algorithms.- A Hierarchy of Conditions for Asynchronous Interactive Consistency.- Associative Parallel Algorithms for Dynamic Edge Update of Minimum Spanning Trees.- The Renaming Problem as an Introduction to Structures for Wait-Free Computing.- Graph Partitioning in Scientific Simulations: Multilevel Schemes versus Space-Filling Curves.- Process Algebraic Model of Superscalar Processor Programs for Instruction Level Timing Analysis.- Software.- Optimization of the Communications between Processors in a General Parallel Computing Approach Using the Selected Data Technique.- Load Imbalance in Parallel Programs.- Software Carry-Save: A Case Study for Instruction-Level Parallelism.- A Polymorphic Type System for Bulk Synchronous Parallel ML.- Towards an Efficient Functional Implementation of the NAS Benchmark FT.- Asynchronous Parallel Programming Language Based on the Microsoft .NET Platform.- A Fast Pipelined Parallel Ray Casting Algorithm Using Advanced Space Leaping Method.- Formal Modeling for a Real-Time Scheduler and Schedulability Analysis.- Disk I/O Performance Forecast Using Basic Prediction Techniques for Grid Computing.- Glosim: Global System Image for Cluster Computing.- Exploiting Locality in Program Graphs.- Asynchronous Timed Multimedia Environments Based on the Coordination Paradigm.- Component-Based Development of Dynamic Workflow Systems Using the Coordination Paradigm.- A Multi-threaded Asynchronous Language.- An Efficient Marshaling Framework for Distributed Systems.- Deciding Optimal Information Dispersal for Parallel Computing with Failures.- Parallel Unsupervised k-Windows: An Efficient Parallel Clustering Algorithm.- Applications.- Analysis of Architecture and Design of Linear Algebra Kernels for Superscalar Processors.- Numerical Simulation of Self-Organisation in Gravitationally Unstable Media on Supercomputers.- Communication-Efficient Parallel Gaussian Elimination.- Alternative Parallelization Strategies in EST Clustering.- Protective Laminar Composites Design Optimisation Using Genetic Algorithm and Parallel Processing.- Tools.- A Prototype Grid System Using Java and RMI.- Design and Implementation of a Cost-Optimal Parallel Tridiagonal System Solver Using Skeletons.- An Extended ANSI C for Multimedia Processing.- The Parallel Debugging Architecture in the Intel® Debugger.- Retargetable and Tuneable Code Generation for High Performance DSP.- The Instruction Register File.- A High Performance and Low Cost Cluster-Based E-mail System.- The Presentation of Information in mpC Workshop Parallel Debugger.- Grid-Based Parallel and Distributed Simulation Environment.- Distributed Object-Oriented Web-Based Simulation.- GEPARD – General Parallel Debugger for MVS-1000/M.- Development of Distributed Simulation System.- CMDE: A Channel Memory Based Dynamic Environment for Fault-Tolerant Message Passing Based on MPICH-V Architecture.- DAxML: A Program for Distributed Computation of Phylogenetic Trees Based on Load Managed CORBA.- D-SAB: A Sparse Matrix Benchmark Suite.- DOVE-G: Design and Implementation of Distributed Object-Oriented Virtual Environment on Grid.