MPICH-1

From HP-SEE Wiki

Contents 1 Authors/Maintainers 2 Summary 3 Features 4 Architectural/Functional Overview 5 Usage Overview 6 Dependencies 7 HP-SEE Applications 8 Resource Centers 9 Usage by Other Projects and Communities 10 Recommendations for Configuration and Usage

• Web site: http://www.mcs.anl.gov/research/projects/mpi/mpich1-old/
• Described version: 1.2.7p1
• Licensing: http://www.mcs.anl.gov/research/projects/mpich2/downloads/license.txt
• User documentation: http://www.mcs.anl.gov/research/projects/mpi/www/index.html, Documentation folder within the MPICH-1 distribution archive
• Download: General MPI download page at MCA ANL
• Source code: MPICH-1 download

Authors/Maintainers

Mathematics and Computer Science Division, Argonne National Laboratory (http://www.mcs.anl.gov/).

Summary

MPICH is an open-source, portable implementation of the full MPI-1 standard specification for a wide variety of parallel computing environments, including workstation clusters and massively parallel processors (MPPs). Besides complete implementation of version 1.2 of the MPI Standard, MPICH also implements significant parts of MPI-2, particularly in the area of parallel I/O. The only major feature supported in MPICH but not in MPICH2 is heterogeneity; that is, support for a single MPI application running on systems that use different data representations (e.g., a mixture of IA32 and IA64 systems).

Although discontinued for some time, MPICH implementation (with the last version 1.2.7) is one of the most frequently used MPI library packages. It is widely used on all types of computing resources: small clusters, Grid sites and large-scale HPC facilities. It is supported on wide range of currently present architectures.

MPICH is deployed on target system through standard configure/make process and it supports C, C++, F77, and F90. MPICH provides an abstraction layer for interprocess communication regardless of the operating environment, be it a single massively parallelized shared-memory machine or a cluster of distributed machines. The different environments use different devices for communication. Each device is an underlying transport layer for the message passing. Examples of devices (or so called channels) are p4 for clusters of machines on normal TCP/IP networks and shmem for shared memory SMP machines.

Features

• Listed features

Architectural/Functional Overview

• high level design info, how it works, performance - may be a link, or several links

Usage Overview

MPICH examples

Dependencies

n/a

HP-SEE Applications

Resource Centers

• Resource centre PARADOX
• Resource centre BG MVAPICH-1
• Resource centre HPCG MVAPICH-1
• Resource centre NIIFI SC PVM3, lammpi
• Resource centre IFIN-HH MPICH‐MX

Usage by Other Projects and Communities

• If any

Recommendations for Configuration and Usage