scispace - formally typeset
Search or ask a question

Showing papers by "Steven J. Plimpton published in 1991"


Proceedings ArticleDOI
01 Dec 1991
TL;DR: An overview of the PGA algorithm is provided, highlighting opportunities for significant speed-up in parallel architectures and implementation details and timing results are provided.
Abstract: The iterative phase gradient autofocus (PGA) algorithm for automatically focusing synthetic aperture radar images has been implemented on both a 16384-processor Connection Machine and the 1024-processor nCUBE 2 hypercube. Massive parallelism has proven its value by dramatically reducing processing times over those achieved on sequential machines by an order of magnitude or more. This is especially important for very large images or where high volumes of input and output are encountered. We provide an overview of the PGA algorithm, highlighting opportunities for significant speed-up in parallel architectures. This is followed by implementation details and timing results.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

4 citations


01 Jan 1991
TL;DR: In this article, a parallel implementation of a general Monte Carlo simulation which allows for body-fitted grids, particle weighting, and a variety of surface and flow chemistry models is described, and the authors compare its performance on a 1024-node nCUBE 2 to a serial version for the CRAY-YMP.
Abstract: Direct simulation Monte Carlo is a well-established technique for modeling low density fluid flows. The parallel implementation of a general simulation which allows for body-fitted grids, particle weighting, and a variety of surface and flow chemistry models is described. The authors compare its performance on a 1024-node nCUBE 2 to a serial version for the CRAY-YMP. Experiences with load-balancing the computation via graph-based heuristics and the newer spectral techniques are also discussed. This is a critical issue, since density fluctuations can create orders-of-magnitude differences in computational loads as the simulation progresses. >

2 citations