/pdf/a-bibliography-of-publications-about-the-fortran-programming-q3kg2hu189.pdf

A Bibliography of Publications about the Fortran Programming Language: Part 1: 1956{1980

Two major strategies exist for the exploitation of parallel execution on shared memory multiprocessors. Multiple threads of execution can be created within a shared program context and loop iterations can be scheduled between processors. While the two techniques are well established in isolation, the design of large, general purpose multi-user multiprocessors requires a seamless blend. The authors examine a set of system structures which combine the two parallel computing models to exploit inherent program parallelism. The objective is a general purpose parallel programming environment in which programs can create as many threads as desired, which are scheduled on the share of processor resource allocated to the program. From within any of these threads, the self scheduling of iterations interacts with the thread management so that free processor resource is dynamically allocated to parallel iterations. >

Self scheduling and execution threads

The execution time of Pascal-like programs can be decreased by putting solutions to problems in their maximally parallel forms. In this study Pascal-like programs are considered, and analysis of parallelism is performed by using a dynamic recent data dependency graph (DDG). The means taken in this paper is able to derive maximally parallel versions of the programs and to minimize the execution time. The advantage of this means is not only to save the effort in implementation but also to possess generality and high-efficiency resulting from dynamically locating parallelism in program. The study shows that the maximally parallel program can run in considerably less time than that needed to run the original sequential Pascal-like program.

S. M. Nassar

Papers

Analysis of parallelism in nested DO loops