Introduction to parallel computing : [a practical guide with examples in C] /

Parallelism in computing occurs on many levels--architectural, organizational, network, and algorithmic. This book is unique in that it provides a balanced treatment of the concepts of parallelism on all levels. Following a concise summary of basic computer architecture, Petersen and Arbenz (both, ETHZ, Zurich) discuss parallelism on the algorithmic level. Then instruction level parallelism through loop unrolling, pipelining, and vectorizing are treated, with examples drawn from popular microprocessors. Shared memory parallelism is discussed next, with examples drawn from supercomputers. Lastly, network parallelism through message passing is explained. This book is unique in that both theoretical concepts and practical examples are plainly presented and clearly discussed. Many examples are also illustrated with program segments written in C. This book, a tutorial, carefully compiles practical tricks and methods that have been successfully used in building fast machines and supercomputers. Hence, it emphasizes what can be done given a real problem. For computer science undergraduates learning about parallelism and for those who develop programs and systems that exploit as much parallelism as possible so as to maximize the desired performance. ^BSumming Up: Highly recommended. Upper-division undergraduates through professionals. J. Y. Cheung University of Oklahoma