Open Access
Techniques for designing efficient parallel programs
P Pieter Struik
- 01 Jan 1991
- Vol. 9132
TL;DR: This paper presents techniques for designing parallel programs, i.e. starting from a formal specification of the problem, which offer the possibility of tuning to the characteristics of the machine on which it is executed and demonstrates two techniques for designs of a parameterized grain size.
read more
Abstract: In this paper we present techniques for designing parallel programs. These techniques are calculational, i.e. starting from a formal specification of the problem we design a program by transforming the specification in a number of steps. The programs that we obtain are correct by design. We demonstrate two techniques by means of an example. First, we illustrate the derivation of a fine grained program. Since the communication overhead of such a program is too large, it can not be implemented on a processor network (e.g. transputer network) efficiently. We, subsequently, demonstrate two techniques for designing programs of a parameterized grain size. For obtaining efficient execution, such programs offer the possibility of tuning to the characteristics of the machine on which it is executed. Finally, we give a complexity analysis of the techniques presented and compare the results from the analysis to experimental results obtained from a transputer implementation. keywords: design techniques, parallel programming, grain size
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
•Proceedings Article
Concrete process algebra
Jos C. M. Baeten,Chris Verhoef +1 more
- 15 Jun 1995
TL;DR: Derivation rules for the encapsulation operator are described, which describe the mechanism for abstracting two atomic actions, say a and b, from each other.
Automatic generation of parallel programs with dynamic load balancing for a network of workstations
Bruce S. Siegell
- 20 Nov 1995
TL;DR: In this article, the authors describe a system that supports dynamic load balancing of distributed applications consisting of parallelized DOALL and DOACROSS loops and demonstrate how parameters for load balancing can be selected and controlled automatically at run time with cooperation between the compiler and runtime system.
50
Real space process algebra
Jos C. M. Baeten,Jan A. Bergstra,Jan A. Bergstra +2 more
- 01 Nov 1993
TL;DR: The overall emphasis is on the introductory explanation of the features of real space process algebra, and characteristic examples are given for each of these.
Real Space Process Algebra
Jos C. M. Baeten,Jan A. Bergstra,Jan A. Bergstra +2 more
- 26 Aug 1991
TL;DR: In this article, the authors extend the real time process algebra of [BB91] to real space-time process algebra, where actions are not just parametrized by a time coordinate, but also by three spatial coordinates.
The modelling and analysis of queueing systems with QNM-ExSpect
W.M.P. van der Aalst
- 01 Jan 1991
TL;DR: The application of ExSpect to Queueing Networks (QN) is described and how to translate a model in QNM-ExSpect to a BCMP network is shown.
References
The VLSI-programming language Tangram and its translation into handshake circuits
Kees van Berkel,Joep L. W. Kessels,Marly Roncken,Ronald W. J. J. Saeijs,F. Schalij +4 more
- 25 Feb 1991
TL;DR: A silicon compiler is constructed that automates the translation of Tangram programs into handshake circuits and converts these handshake circuits into asynchronous circuits and subsequently into VLSI layouts.
224
•Book
Parallel Computations
G. Rodrigue
- 01 Jan 1982
TL;DR: The final author version and the galley proof are versions of the publication after peer review and the final published version features the final layout of the paper including the volume, issue and page numbers.
129
Parallel programs for the recognition of P-invariant segments
J.P. Katoen,L.A.M. Schoenmakers +1 more
- 01 Jan 1991
TL;DR: In a longitudinally moveable antifriction bearing having an endless row of rolling elements, with a zone for loadedrolling elements, a return zone for unloaded rolling elements and reversing sections joining the respective loaded zones and return zones, flexurally elastic tongues are provided.
60
Trace Theory and Systolic Computations
Martin Rem
- 15 Jun 1987
TL;DR: A class of concurrent computations, or special-purpose computing engines, that may be characterized by regular arrangements of simple cells, where the arrangement consumes streams of input values and produces streams of output values and the communication behaviors are independent of the values communicated.
Foundation of compositional program refinement—safety properties
Rob Gerth
- 01 May 1990
TL;DR: This paper develops a foundation for refinement of parallel programs that may synchronously communicate and/or share variables; programs rendered as 1st order transition systems and shows that they yield assertional methods for refinement that resemble the methods used in Z. Manna and A. Pnueli's temporal logic proof system.
40