Recent Advances in Parallel Virtual Machine and Message Passing Interface: 13th European Pvm/Mpi User's Group Meeting, Bonn, Germany, September 17-20, 2006: Proceedings

by Mohr, Bernd; Traff, Jesper Larsson; Worringen, Joachim; Dongarra, Jack J.

• ISBN13: 9783540391104
• ISBN10: 354039110X
• Format: Paperback
• Copyright: 2006-10-15
• Publisher: Springer Verlag

Summary

This book constitutes the refereed proceedings of the 13th European PVM/MPI Users' Group Meeting held in Bonn, Germany in September 2006. The 38 revised full papers presented together with abstracts of 6 invited contributions, 4 tutorial papers and 6 poster papers were carefully reviewed and selected from 84 submissions. The papers are organized in topical sections on collective communication, communication protocols, debugging and verification, fault tolerance, metacomputing and grid, parallel I/O, implementation issues, object-oriented message passing, limitations and extensions, performance, and are completed with 6 contributions to the special ParSim session on current trends in numerical simulation for parallel engineering environments.

Table of Contents

Too big for MPI?	p. 1
Approaches for parallel applications fault tolerance	p. 2
Where does MPI need to grow?	p. 3
Peta-scale supercomputer project in Japan and challenges to life and human simulation in Japan	p. 4
Resource and application adaptivity in message passing systems	p. 5
Performance advantages of partitioned global address space languages	p. 6
Using MPI-2 : a problem-based approach	p. 7
Performance tools for parallel programming	p. 8
High-performance parallel I/O	p. 10
Hybrid MPI and OpenMP parallel programming	p. 11
Issues in developing a thread-safe MPI implementation	p. 12
Scalable parallel suffix array construction	p. 22
Formal verification of programs that use MPI one-sided communication	p. 30
MPI collective algorithm selection and quadtree encoding	p. 40
Parallel prefix (scan) algorithms for MPI	p. 49
Efficient Allgather for regular SMP-clusters	p. 58
Efficient shared memory and RDMA based design for MPIöAllagather over InfiniBand	p. 66
High performance RDMA protocols in HPC	p. 76
Implementation and shared-memory evaluation of MPICH2 over the nemesis communication subsystem	p. 86
MPI/CTP : a reconfigurable MPI for HPC applications	p. 96
Correctness checking of MPI one-sided communication using Marmot	p. 105
An interface to support the identification of dynamic MPI 2 processes for scalable parallel debugging	p. 115
Modeling and verification of MPI based distributed software	p. 123
FT-MPI, fault-tolerant metacomputing and generic name services : a case study	p. 133
Scalable fault tolerant protocol for parallel runtime environments	p. 141
An intelligent management of fault tolerance in cluster using RADICMPI	p. 150
Extended mpiJava for distributed checkpointing and recovery	p. 158
Running PVM applications on multidomain clusters	p. 166
Reliable orchestration of distributed MPI-applications in a UNICORE-based grid with MetaMPICH and MetaScheduling	p. 174
The new multidevice architecture of MetaMPICH in the context of other approaches to grid-enabled MPI	p. 184
Using an enterprise grid for execution of MPI parallel applications - a case study	p. 194
Self-adaptive hints for collective I/O	p. 202
Exploiting shared memory to improve parallel I/O performance	p. 212
High-bandwidth remote parallel I/O with the distributed memory filesystem MEMFS	p. 222
Effective seamless remote MPI-I/O operations with derived data types using PVFS2	p. 230
Automatic memory optimizations for improving MPI derived datatype performance	p. 238
Improving the dynamic creation of processes in MPI-2	p. 247
Non-blocking Java communications support on clusters	p. 256
Modernizing the C++ interface to MPI	p. 266
Can MPI be used for persistent parallel services?	p. 275
Observations on MPI-2 support for hybrid master/slave applications in dynamic and heterogeneous environments	p. 285
What MPI could (and cannot) do for mesh-partitioning on non-homogeneous networks	p. 293
Scalable parallel trace-based performance analysis	p. 303
TAUg : runtime global performance data access using MPI	p. 313
Tracing the MPI-IO calls' disk accesses	p. 322
Measuring MPI send and receive overhead and application availability in high performance network interfaces	p. 331
Challenges and issues in benchamarking MPI	p. 339
Implementation and usage of the PERUSE-interface in open MPI	p. 347
Current trends in numerical simulation for parallel engineering environments	p. 356
MPJ express meets gadget : towards a Java code for cosmological simulations	p. 358
An approach for parallel fluid-structure interaction on unstructured meshes	p. 366
Optimizing a conjugate gradient solver with non-blocking collective operations	p. 374
Parallel DSMC gasflow simulation of an in-line coater for reactive sputtering	p. 383
Parallel simulation of T-M processes in underground repository of spent nuclear fuel	p. 391
On the usability of high-level parallel IO in unstructured grid simulations	p. 400
Automated performance comparison	p. 402
Improved GROMACS scaling on ethernet switched clusters	p. 404
Asynchronity in collective operation implementation	p. 406
PARUS : a parallel programming framework for heterogeneous multiprocessor systems	p. 408
Application of PVM to protein homology search	p. 410
Table of Contents provided by Blackwell. All Rights Reserved.