Parallel Processing Of Discrete Problems

In the past two decades, breakthroughs in computer technology have made a tremendous impact on optimization. In particular, availability of parallel computers has created substantial interest in exploring the use of parallel processing for solving discrete and global optimization problems. The chapters in this volume cover a broad spectrum of recent research in parallel processing of discrete and related problems. The topics discussed include distributed branch-and-bound algorithms, parallel genetic algorithms for large scale discrete problems, simulated annealing, parallel branch-and-bound search under limited-memory constraints, parallelization of greedy randomized adaptive search procedures, parallel optical models of computing, randomized parallel algorithms, general techniques for the design of parallel discrete algorithms, parallel algorithms for the solution of quadratic assignment and satisfiability problems. The book will be a valuable source of information to faculty, students and researchers in combinatorial optimization and related areas.

This book contains papers presented at the Workshop on Parallel Processing of Discrete Optimization Problems held at DIMACS in April 1994. The contents cover a wide spectrum of the most recent algorithms and applications in parallel processing of discrete optimization and related problems. Topics include parallel branch and bound algorithms, scalability, load balancing, parallelism and irregular data structures and scheduling task graphs on parallel machines. Applications include parallel algorithms for solving satisfiability problems, location problems, linear programming, quadratic and linear assignment problems. This book would be suitable as a textbook in advanced courses on parallel algorithms and combinatorial optimization.

Parallel processing has been an enabling technology in scientific computing for more than 20 years. This book is the first in-depth discussion of parallel computing in 10 years; it reflects the mix of topics that mathematicians, computer scientists, and computational scientists focus on to make parallel processing effective for scientific problems. Presently, the impact of parallel processing on scientific computing varies greatly across disciplines, but it plays a vital role in most problem domains and is absolutely essential in many of them. Parallel Processing for Scientific Computing is divided into four parts: The first concerns performance modeling, analysis, and optimization; the second focuses on parallel algorithms and software for an array of problems common to many modeling and simulation applications; the third emphasizes tools and environments that can ease and enhance the process of application development; and the fourth provides a sampling of applications that require parallel computing for scaling to solve larger and realistic models that can advance science and engineering.

Efficient parallel solutions have been found to many problems. Some of them can be obtained automatically from sequential programs, using compilers. However, there is a large class of problems - irregular problems - that lack efficient solutions. IRREGULAR 94 - a workshop and summer school organized in Geneva - addressed the problems associated with the derivation of efficient solutions to irregular problems. This book, which is based on the workshop, draws on the contributions of outstanding scientists to present the state of the art in irregular problems, covering aspects ranging from scientific computing, discrete optimization, and automatic extraction of parallelism. Audience: This first book on parallel algorithms for irregular problems is of interest to advanced graduate students and researchers in parallel computer science.

This book offers readers a set of new approaches and tools a set of tools and techniques for facing challenges in parallelization with design of embedded systems. It provides an advanced parallel simulation infrastructure for efficient and effective system-level model validation and development so as to build better products in less time. Since parallel discrete event simulation (PDES) has the potential to exploit the underlying parallel computational capability in today’s multi-core simulation hosts, the author begins by reviewing the parallelization of discrete event simulation, identifying problems and solutions. She then describes out-of-order parallel discrete event simulation (OoO PDES), a novel approach for efficient validation of system-level designs by aggressively exploiting the parallel capabilities of todays’ multi-core PCs. This approach enables readers to design simulators that can fully exploit the parallel processing capability of the multi-core system to achieve fast speed simulation, without loss of simulation and timing accuracy. Based on this parallel simulation infrastructure, the author further describes automatic approaches that help the designer quickly to narrow down the debugging targets in faulty ESL models with parallelism.

Mathematics of Computing -- Parallelism.