This volume contains most of the papers presented at theworkshop on research directions in high-level parallelprogramming languages, held at Mont Saint-Michel, France, inJune 1991. The motivation for organizing this workshop camefrom the emergence of a new class of formalisms fordescribing parallel computations in the last few years.Linda, Unity, Gamma, and the Cham are the most significantrepresentatives of this new class. Formalisms of this familypromote simple but powerful language features for describingdata and programs. These proposals appeared in differentcontexts and were applied in different domains, and the goalof the workshop was to review the status of this new fieldand compare experiences. The workshop was organized intofour main sessions: Unity, Linda, Gamma, and ParallelProgram Design. The corresponding parts ofthe volume areintroduced respectively by J. Misra, D. Gelernter, D. LeM tayer, and J.-P. Ban tre.
A perspective on parallel program design.- UNITY to UC: A case study in the derivation of parallel programs.- Reasoning about synchronic groups.- An industrial experience in the use of UNITY.- On the UNITY design decisions.- Flexible program structures for concurrent programming.- Current research on Linda.- Expressing fine-grained parallelism using concurrent data structures.- Persistent Linda: Linda + transactions + query processing.- Parallel logic programming using the Linda model of computation.- Lucinda - A polymorphic Linda.- A rationale for programming with Ease.- Intermediate uniformly distributed tuple space on transputer meshes.- Mixing concurrency abstractions and classes.- Coordination applications of Linda.- The chemical reaction model.- to Gamma.- Parallel programming with bags.- Implementation of Gamma on the connection machine.- Parallel programming with pure functional languages.- Parallel programming in maude.- Parallel program design.- The palindrome systolic array revisited.- The synthesis of systolic programs.- Synthesizing delay insensitive circuits from verified programs.- A distributed implementation of a task pool.- Invariance and contraction by infinite iterations of relations.- Constructing a parallel sort program on hypercube by using Distributively Callable Procedures.
Series: Lecture Notes in Computer Science
Number Of Pages: 392
Published: 12th February 1992
Publisher: Springer-Verlag Berlin and Heidelberg Gmbh & Co. Kg
Country of Publication: DE
Dimensions (cm): 23.39 x 15.6
Weight (kg): 0.56