DARPA Parallel Architecture Benchmark Study

Christopher M. Brown, Robert Fowler, Thomas J. LeBlanc, Michael L. Scott, Mandayam Srinivas, Liudvikas Bukys, John Costanzo, Lawrence A. Crowl, Peter Dibble, Neal Gafter, Brian Marsh, Thomas Olson, and Laura Sanchis, "DARPA Parallel Architecture Benchmark Study", Butterfly Project Report 13, Department of Computer Science, University of Rochester, Rochester, New York, 14627-0226, October 1986

In intensive work over a four-week period in the summer of 1986, seven problems were studied and implemented on the Butterfly. The problems were inspired by various problems in computer vision, and were proposed as benchmarks for a DARPA workshop on parallel architectures. They were: convolution and zero-crossing detection for edges, edge tracking, connected component labeling, hough transform, three computational geometry problems (convex hull, voronoi diagram, and minimal spanning tree), three-dimensional visibility calculations, subgraph isomorphism, and minimum cost path calculation. BPRs 10, 11, and 14 are detailed reports on three of the problems. BPR 13 contains the conclusions of the study and writeups of the work not covered in other BPRs.


Comments to Lawrence@Crowl.org.
Last modified on 02 Feb 1900.