Regular article
Paragon: A parallel programming environment for scientific applications using communication structures

https://doi.org/10.1016/0743-7315(92)90025-IGet rights and content
Under a Creative Commons license
open archive

Abstract

The Paragon project is directed toward the identification and exploration of programming methodologies to be used for the development of large-scale scientific applications on modern parallel computers. The Paragon environment consists of a set of data-parallel programming constructs and a flexible run-time environment. Among the salient features of Paragon are a data-parallel programming paradigm that is applicable to a wide variety of high-performance architectures, a practical abstraction of data distribution through the shape construct, and support for a powerful style of programming using communication structures. In this paper, programming with communication structures is considered and the performance of Paragon is characterized both for primitive operations and for several applications.

Cited by (0)

This research was conducted using the Cornell National Supercomputer Facility, a resource of the Center for Theory and Simulation in Science and Engineering at Cornell University, which receives major funding from the National Science Foundation and IBM Corp., with additional support from New York State and members of its Corporate Research Institute.