An intensional parallel processing language for applications programming:

Abstract: "Parallel programming languages fall into two classes, the implicitly parallel langugaes and the explicitly parallel languages. We will consider the implicitly parallel languages. The explicitly parallel languages are important in that they are motivated by implementation consideratio...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Ashcroft, Edward A. (VerfasserIn), Faustini, A. A. (VerfasserIn), Jagannathan, Rangaswamy (VerfasserIn)
Format: Buch
Sprache:English
Veröffentlicht: Menlo Park, Calif. 1989
Schriftenreihe:Computer Science Laboratory <Menlo Park, Calif.>: SRI-CSL 89,1
Schlagworte:
Zusammenfassung:Abstract: "Parallel programming languages fall into two classes, the implicitly parallel langugaes and the explicitly parallel languages. We will consider the implicitly parallel languages. The explicitly parallel languages are important in that they are motivated by implementation considerations. Compiler technology is relatively simple in this context. As a result we see a rapid proliferation of explicitly parallel languages on multi-processor architectures. This has important consequences for parallel processing in the short-term. The implicitly parallel languages are equally important in that they favor the applications software. The software developed is more portable and maintainable over a software life cycle that can easily extend over successive generations of multi-processor architectures
This has important consequences for the long-term success of parallel processing. Languages that are not based on a sequential model of computation can exhibit significant amounts of implicit parallelism. The declaritive programming languages are therefore important parallel programming languages. Within the declaritive paradigm we feel that the languages based on intensional logic are unique in that they permit data structures such are arrays, lists and trees to be implemented in a manner that is easily distributed. To illustrate the concepts we have chosen a particular language, namely, Lucid. However the same principles apply to other implicitly parallel languages
Lucid's simplicity and semantic elegance make possible the inclusion of a non-trivial application program -- an adaptive solution to the n-body problem. We present an analysis of the implicit parallelism available in such a program and show ways in which we can harness this parallelism.
Beschreibung:27 S. graph. Darst.

Es ist kein Print-Exemplar vorhanden.

Fernleihe Bestellen Achtung: Nicht im THWS-Bestand!