Parallel Algorithms for Regular Architectures – Meshes & Pyramids: Meshes and Pyramids (Scientific Computation) - Hardcover

Miller, Russ

Hardcover

ISBN 10: 0262132338 ISBN 13: 9780262132336

Publisher: MIT Press, 1996

Synopsis

Parallel-Algorithms for Regular Architectures is the first book to concentrate exclusively on algorithms and paradigms for programming parallel computers such as the hypercube, mesh, pyramid, and mesh-of-trees. Algorithms are given to solve fundamental tasks such as sorting and matrix operations, as well as problems in the field of image processing, graph theory, and computational geometry. The first chapter defines the computer models, problems to be solved, and notation that will be used throughout the book. It also describes fundamental abstract data movement operations that serve as the foundation to many of the algorithms presented in the book. The remaining chapters describe efficient implementations of these operations for specific models of computation and present algorithms (with asymptotic analyses) that are often based on these operations. The algorithms presented are the most efficient known, including a number of new algorithms for the hypercube and mesh-of-trees that are better than those that have previously appeared in the literature. The chapters may be read independently, allowing anyone interested in a specific model to read the introduction and then move directly to the chapter(s) devoted to the particular model of interest. Russ Miller is Assistant Professor in the Department of Computer Science, State University of New York at Buffalo. Quentin F. Stout is Associate Professor in the Department of Electrical Engineering and Computer Science at the University of Michigan. Parallel Algorithms for Regular Architectures is included in the Scientific Computation series, edited by Dennis Gannon.

About this title

Synopsis

"Parallel Algorithms for Regular Architectures" offers a collection of optimal and efficient algorithms for solving problems on sets of processors configured as a mesh or a pyramid. In addition, it provides insights into the parallel algorithm design process. The algorithms can be adapted to multiprocessor machines with varying degrees of granularity and a variety of processor configurations. They can be utilized inside a single VLSI chip, on special-purpose parallel machines, on intermediate-size machines, or on large parallel computers, such as the IBM SP2, Cray T3D/T3E, Intel Paragon, or MasPar MP1/MP2, that are routinely used for solving complex scientific and engineering problems. Basic algorithms, such as sorting, matrix multiplication and parallel prefix, are described, as are algorithms to solve fundamental problems in image processing, computational geometry and graph theory. A consistent approach, based on algorithmic techniques, shows how to exploit the use of data movement operations.

Since many of the algorithms were originally created by the authors using these techniques, the reader can actually see how parallel algorithms are developed by following the design process. "Parallel Algorithms for Regular Architectures" should be useful to researchers as well as practitioners who need to implement parallel programmes. The algorithms and operations can be incorporated into a variety of applications, and the design and analysis techniques can be exploited in an even greater range.

About the Author

Russ Miller is Assistant Professor in the Department of Computer Science, State University of New York at Buffalo. Quentin F. Stout is Associate Professor in the Department of Electrical Engineering and Computer Science at the University of Michigan.

Edition details

• Publisher: MIT Press
• Publication date: 1996
• Language: English
• ISBN 10: 0262132338
• ISBN 13: 9780262132336
• Binding: Hardcover
• Number of pages: 328