Synopses & Reviews
With recent advances in VLSI technology, parallel computers are becoming more prevalent and affordable. The SIMD Model of Parallel Computation presents parallel architecture and parallel algorithms to non-specialists in computer science and engineering. Beginning with an overview of a variety of parallel architectures, the book details the interaction between parallel architectures and parallel algorithms by presenting, for each architecture, algorithms that are tailored to the given architecture. The focus is on the SIMD (Single Instruction Stream, Multiple Data Stream) model of parallel computation and its implementation on both SIMD and MIMD (Multiple Instruction Stream, Multiple Data Stream) architectures. The book presents both introductory and advanced material in parallel computation and for each topology, several existing and proposed parallel machines are discussed and compared. Included is an analysis of parallel algorithms for image processing, scientific and symbolic tasks. The effects of architectural decisions on algorithms are also examined in detail.
Synopsis
1.1 Background There are many paradigmatic statements in the literature claiming that this is the decade of parallel computation. A great deal of research is being de voted to developing architectures and algorithms for parallel machines with thousands, or even millions, of processors. Such massively parallel computers have been made feasible by advances in VLSI (very large scale integration) technology. In fact, a number of computers having over one thousand pro cessors are commercially available. Furthermore, it is reasonable to expect that as VLSI technology continues to improve, massively parallel computers will become increasingly affordable and common. However, despite the significant progress made in the field, many funda mental issues still remain unresolved. One of the most significant of these is the issue of a general purpose parallel architecture. There is currently a huge variety of parallel architectures that are either being built or proposed. The problem is whether a single parallel computer can perform efficiently on all computing applications."