Recent Advances in Parallel Virtual Machine and Message Passing Interface

Themessagepassingparadigmisconsideredthemoste?ectivewaytodevelop- ?cient parallel applications. PVM (Parallel Virtual Machine) and MPI (Message Passing Interface) are the most frequently used tools for programming message passing applications. This volume includes the selected contributions presented at the 10th - ropean PVM/MPI Users’ Group Meeting (Euro PVM/MPI 2003), which was held in Venice, Italy, September 29–October 2, 2003. The conference was jointly organized by the Department of Computer Science of the Ca’ Foscari University of Venice, Italy and the Information Science and Technologies Institute of the National Research Council (ISTI-CNR), Pisa, Italy. TheconferencewaspreviouslyheldinLinz,Austria(2002),Santorini,Greece (2001), Balatonfured, ¨ Hungary (2000), Barcelona, Spain (1999), Liverpool, UK (1998), and Krakow, Poland (1997). The ?rst three conferences were devoted to PVM and were held in Munich, Germany (1996), Lyon, France (1995), and Rome, Italy (1994). The conference has become a forum for users and developers of PVM, MPI, and other message passing environments. Interactions between these groups has proved to be very useful for developing new ideas in parallel computing, and for applying some of those already existent to new practical ?elds. The main topics of the meeting were evaluation and performance of PVM and MPI, ext- sions, implementations and improvements of PVM and MPI, parallel algorithms using the message passing paradigm, and parallel applications in science and engineering. In addition, the topics of the conference were extended to include Grid computing, in order to re?ect the importance of this area for the hi- performance computing community.

Invited Talks.- Messaging Systems: Parallel Computing the Internet and the Grid.- Progress towards Petascale Virtual Machines.- Future Developments in MPI.- Integrating Scalable Process Management into Component-Based Systems Software.- Programming High Performance Applications Using Components.- ASSIST High-Performance Programming Environment: Application Experiences and Grid Evolution.- Tutorials.- High-Level Programming in MPI.- Programming Environments for Grids and Distributed Computing Systems.- Evaluation and Performance Analysis.- Performance Modeling and Evaluation of Java Message-Passing Primitives on a Cluster.- Integrating New Capabilities into NetPIPE.- Off-Line Performance Prediction of Message-Passing Applications on Cluster Systems.- Complexity Driven Performance Analysis.- Usefulness and Usage of SKaMPI-Bench.- The Performance of Parallel Disk Write Methods for Linux Multiprocessor Nodes.- A Model for Performance Analysis of MPI Applications on Terascale Systems.- Evaluating the Performance of MPI-2 Dynamic Communicators and One-Sided Communication.- Ring Algorithms on Heterogeneous Clusters with PVM: Performance Analysis and Modeling.- An MPI Tool to Measure Application Sensitivity to Variation in Communication Parameters.- Measuring MPI Latency Variance.- Parallel Algorithms Using Message Passing.- CGMgraph/CGMlib: Implementing and Testing CGM Graph Algorithms on PC Clusters.- Efficient Parallel Implementation of Transitive Closure of Digraphs.- A Scalable Crystallographic FFT.- Object-Oriented NeuroSys: Parallel Programs for Simulating Large Networks of Biologically Accurate Neurons.- PageRank Computation Using PC Cluster.- An Online Parallel Algorithm for Remote Visualization of Isosurfaces.- Parallel Algorithms for Computing the Smith Normal Form of Large Matrices.- Hierarchical MPI+OpenMP Implementation of Parallel PIC Applications on Clusters of Symmetric MultiProcessors.- Non-strict Evaluation of the FFT Algorithm in Distributed Memory Systems.- A Parallel Approach for the Solution of Non-Markovian Petri Nets.- Advanced Hybrid MPI/OpenMP Parallelization Paradigms for Nested Loop Algorithms onto Clusters of SMPs.- The AGEB Algorithm for Solving the Heat Equation in Two Space Dimensions and Its Parallelization on a Distributed Memory Machine.- A Parallel Scheme for Solving a Tridiagonal Matrix with Pre-propagation.- Competitive Semantic Tree Theorem Prover with Resolutions.- Explicit Group Iterative Solver on a Message Passing Environment.- Applying Load Balancing in Data Parallel Applications Using DASUD.- Performance Analysis of Approximate String Searching Implementations for Heterogeneous Computing Platform.- Extensions, Improvements and Implementations of PVM/MPI.- Using a Self-connected Gigabit Ethernet Adapter as a memcpy() Low-Overhead Engine for MPI.- Improving the Performance of Collective Operations in MPICH.- PVMWebCluster: Integration of PVM Clusters Using Web Services and CORBA.- Lock-Free Collective Operations.- Efficient Message-Passing within SMP Systems.- The Network Agnostic MPI – Scali MPI Connect.- PC/MPI: Design and Implementation of a Portable MPI Checkpointer.- Improving Generic Non-contiguous File Access for MPI-IO.- Remote Exception Handling for PVM Processes.- Evaluation of an Eager Protocol Optimization for MPI.- A Comparison of MPICH Allgather Algorithms on Switched Networks.- Network Fault Tolerance in LA-MPI.- MPI on BlueGene/L: Designing an Efficient General Purpose Messaging Solution for a Large Cellular System.- Porting P4 to Digital Signal Processing Platforms.- Fast and Scalable Barrier Using RDMA and Multicast Mechanisms for InfiniBand-Based Clusters.- A Component Architecture for LAM/MPI.- ORNL-RSH Package and Windows ‘03 PVM 3.4.- MPI for the Clint Gb/s Interconnect.- Implementing Fast and Reusable Datatype Processing.- An MPI Implementation Supported by Process Migration and Load Balancing.- PVM over the CLAN Network.- Parallel Programming Tools.- Distributed Configurable Application Monitoring on SMP Clusters.- Integrating Multiple Implementations and Structure Exploitation in the Component-Based Design of Parallel ODE Solvers.- Architecture of Monitoring System for Distributed Java Applications.- A Communication API for Implementing Irregular Algorithms on SMP Clusters.- TOM – Efficient Monitoring Infrastructure for Multithreaded Programs.- MPI Farm Programs on Non-dedicated Clusters.- Application Composition in Ensemble Using Intercommunicators and Process Topologies.- Improving Properties of a Parallel Program in ParJava Environment.- Applications in Science and Engineering.- Flow Pattern and Heat Transfer Rate in Three-Dimensional Rayleigh-Benard Convection.- A Parallel Split Operator Method for the Time Dependent Schrödinger Equation.- A Parallel Software for the Reconstruction of Dynamic MRI Sequences.- Improving Wildland Fire Prediction on MPI Clusters.- Building 3D State Spaces of Virtual Environments with a TDS-Based Algorithm.- Parallel Pencil-Beam Redefinition Algorithm.- Dynamic Load Balancing for the Parallel Simulation of Cavitating Flows.- Message Passing Fluids: Molecules as Processes in Parallel Computational Fluids.- Parallel Implementation of Interval Analysis for Equations Solving.- A Parallel System for Performing Colonic Tissue Classification by Means of a Genetic Algorithm.- Eigenanalysis of Finite Element 3D Flow Models by Parallel Jacobi–Davidson.- Grid and Heterogeneous Computing.- Executing and Monitoring PVM Programs in Computational Grids with Jini.- Multiprogramming Level of PVM Jobs in a Non-dedicated Linux NOW.- Mapping and Load-Balancing Iterative Computations on Heterogeneous Clusters.- Dynamic Topology Selection for High Performance MPI in the Grid Environments.- Monitoring Message Passing Applications in the Grid with GRM and R-GMA.- Component-Based System for Grid Application Workflow Composition.- Evaluating and Enhancing the Use of the GridFTP Protocol for Efficient Data Transfer on the Grid.- Resource Monitoring and Management in Metacomputing Environments.- Generating an Efficient Dynamics Multicast Tree under Grid Environment.- Topology-Aware Communication in Wide-Area Message-Passing.- Design and Implementation of Dynamic Process Management for Grid-Enabled MPICH.- Scheduling Tasks Sharing Files on Heterogeneous Clusters.- Special Session: ParSim 03.- Special Session of EuroPVM/MPI 2003: Current Trends in Numerical Simulation for Parallel Engineering Environments ParSim 2003.- Efficient and Easy Parallel Implementation of Large Numerical Simulations.- Toward a Scalable Algorithm for Distributed Computing of Air-Quality Problems.- A Piloting SIMulator for Maritime and Fluvial Navigation: SimNav.- Methods and Experiences of Parallelizing Flood Models.- padfem2 – An Efficient, Comfortable Framework for Massively Parallel FEM-Applications.- AUTOBENCH/AUTO-OPT: Towards an Integrated Construction Environment for Virtual Prototyping in the Automotive Industry.