Researchers and practitioners involved with the development and utilization of parallel languages, software, and systems will find this collection of articles a useful summary of the history, development, programming, and applications of the Distributed Array of Processors (DAP) architecture.
Manufactured by Active Memory Technology in Britain, DAP provides massively parallel, fine grain processing and is an example of a Single Instruction stream Multiple Data stream (SIMD) architecture where multiple processors are under a single control.
The DAP is basically a software computer and has been shown to have a much wider range of applicability than special purpose floating point processors or image processing systems.
Articles in the book's final section demonstrate the breadth of DAP applicability, noting that while applications for the DAP are suitable in a general sense if they are CUP-limited on normal machines and are large enough to take advantage of the many microprocessors available for concurrent processing, there are several secondary characteristics that the DAP might exploit. These include large amounts of logical, character or integer operations; operations over large structured data sets; and applications which can benefit by special data representations - unusually short or long numerical forms, modular arithmetic, fixed point number representions, and character representations.
Dennis Parkinson was formerly Director of the Center for Parallel Computing at Queen Mary College, London. Massively Parallel Computing with the DAP is included in the series Research Monographs in Parallel and Distributed Computing. Copublished with Pitman Publishing.
Manufactured by Active Memory Technology in Britain, DAP provides massively parallel, fine grain processing and is an example of a Single Instruction stream Multiple Data stream (SIMD) architecture where multiple processors are under a single control.
The DAP is basically a software computer and has been shown to have a much wider range of applicability than special purpose floating point processors or image processing systems.
Articles in the book's final section demonstrate the breadth of DAP applicability, noting that while applications for the DAP are suitable in a general sense if they are CUP-limited on normal machines and are large enough to take advantage of the many microprocessors available for concurrent processing, there are several secondary characteristics that the DAP might exploit. These include large amounts of logical, character or integer operations; operations over large structured data sets; and applications which can benefit by special data representations - unusually short or long numerical forms, modular arithmetic, fixed point number representions, and character representations.
Dennis Parkinson was formerly Director of the Center for Parallel Computing at Queen Mary College, London. Massively Parallel Computing with the DAP is included in the series Research Monographs in Parallel and Distributed Computing. Copublished with Pitman Publishing.
