Parallel Computers 2: Architecture, Programming and Algorithms

ISBN-13: 9780367456047 / Angielski / Miękka / 2020 / 642 str.

R. W. Hockney; C. R. Jesshope

Opis

Parallel Computers 2: Architecture, Programming and Algorithms reflects the shift in emphasis of parallel computing and tracks the development of supercomputers in the years since the first edition was published. It looks at large-scale parallelism as found in transputer ensembles. This extensively rewritten second edition includes major new sectio

Recenzje

"… as a comprehensive and readable introduction, this edition is guaranteed as long a life as its predecessor. As all supercomputer manufacturers embrace parallel processing, it will be found more frequently on physicists' bookshelves."
-Contemporary Physics

"…This book continues to be the recommended introduction into parallel computers. It provides a wide view of the machines currently available and the way in which they are progressing. … it represents a worthwhile investment for all interested in the field."
-BCS Parallel Processing Specialist Group

"Anyone interested in high end computing would do well to introduce themselves to the necessary software techniques sooner rather than later. With a bibliography containing over 500 references and a good index, this book is also a thorough reference work and in summary is a welcome addition to the technical computing literature."
-The Australian Physicist

Spis treści

Introduction: History of parallelism and supercomputing. Classification of designs. Characteristics of performance. Pipelined computers: Selection and comparison. The CRAY X-MP and CRAY-2. The CDC CYBER and ETA 10. Japanese vector computers. The FPS AP-120B and derivatives. Multiprocessors and multiprocessor arrays: The limitations of pipelining. The alternative of replication. Switching networks. An historical perspective. Replication - a future with VLSI. Parallel languages: Introduction. Implicit parallelism and vectorisation. Structure parallelism. Process parallelism. Techniques for exploiting parallelism. Parallel algorithms: General principles. Recurrences. Matrix multiplication. Tridiagonal systems. Transforms. Partial differential equations. Technology and the future: Characterisation. Bipolar technologies (TTL, ECL, I2L). MOS technologies (NMOS and CMOS). Scaling technologies. The problem with scaling. System partitioning. Wafer-scale integration. The last word. Appendix. References. Index.

O autorze

Hockney, R.W | Jesshope, C.R