The world of modern computing potentially offers many helpful methods and tools to scientists and engineers, but the fast pace of change in computer hardware, software, and algorithms often makes practical use of the newest computing technology difficult. The Scientific and Engineering Computation series focuses on rapid advances in computing technologies and attempts to facilitate transferring these technologies to applications in science and engineering. It will include books on theories, methods, and original applications in such areas as parallelism, large-scale simulations, time-critical computing, computer-aided design and engineering, use of computers in manufacturing, visualization of scientific data, and human-machine interface technology.
The series will help scientists and engineers to understand the current world of advanced computation and to anticipate future developments that will impact their computing environments and open up new capabilities and modes of computation.
This volume presents a software package for developing parallel programs executable on networked Unix computers. The tool called Parallel Virtual Machine (PVM) allows a heterogeneous collection of workstations and supercomputers to function as a single high-performance parallel machine. PVM is portable and runs on a wide variety of modern platforms. It has been well accepted by the global computing community and used successfully for solving large-scale problems in science, industry, and business.
Janusz S. Kowalik
Preface
In this book we describe the Parallel Virtual Machine (PVM) system and how to develop programs using PVM. PVM is a software system that permits a heterogeneous collection of Unix computers networked together to be viewed by a user's program as a single parallel computer. PVM is the mainstay of the Heterogeneous Network Computing research project, a collaborative venture between Oak Ridge National Laboratory, the University of Tennessee, Emory University, and Carnegie Mellon University.
The PVM system has evolved in the past several years into a viable technology for distributed and parallel processing in a variety of disciplines. PVM supports a straightforward but functionally complete message-passing model.
PVM is designed to link computing resources and provide users with a parallel platform for running their computer applications, irrespective of the number of different computers they use and where the computers are located. When PVM is correctly installed, it is capable of harnessing the combined resources of typically heterogeneous networked computing platforms to deliver high levels of performance and functionality.
In this book, we describe the architecture of the PVM system and discuss its computing model; the programming interface it supports; auxiliary facilities for process groups; the use of PVM on highly parallel systems such as the Intel Paragon, Cray T3D, and Thinking Machines CM-5; and some of the internal implementation techniques employed. Performance issues, dealing primarily with communication overheads, are analyzed, and recent findings as well as enhancements are presented. To demonstrate the viability of PVM for large-scale scientific supercomputing, we also provide some example programs.
This book is not a textbook; rather, it is meant to provide a fast entrance to the world of heterogeneous network computing. We intend this book to be used by two groups of readers: students and researchers working with networks of computers. As such, we hope this book can serve both as a reference and as a supplement to a teaching text on aspects of network computing.
This guide will familiarize readers with the basics of PVM and the concepts used in programming on a network. The information provided here will help with the following PVM tasks: