Experience with distributed programming in Orca

Abstract

Orca is a language for programming parallel applications on distributed computing systems. Although processors in such systems communicate only through message passing and not through shared memory, Orca provides a communication model based on logically shared data. The language and its implementation are briefly described and a report is given on experiences in using Orca for three parallel applications: the traveling salesman problem, the all-pairs shortest paths problem, and successive overrelaxation. These applications have different needs for shared data: TSP benefits greatly from the support for shared data; ASP benefits from the use of broadcast communication, even though it is hidden in the implementation; SOR merely requires point-to-point communication, but still can be implemented in the language by simulating message passing. How these applications are programmed in Orca is discussed and the most interesting portions of the Orca code are given. Performance measurements for these programs on a distributed system consisting of 10 MC68020s connected by an Ethernet are also included. These measurements show that significant speedups are obtained for all three programs.<>