Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 55
APPENDIX B ANNOTATED SHORT BIBLIOGRAPHY OF ALGORITHM- OR APPLICATION-SPECIALIZED COMPUTER SYSTEMS This annotated bibliography lists and briefly discusses some literature on computer systems that have been built to execute one or a small number of computation structures with maximum speed. The common characteristic is parallelism. The common design theme is to create a configuration of processors and interconnections that can execute specific parallel computation structures at full capacity of all of the processors. This research has a long history in some application areas and a much briefer history in others. Architectures for efficient high performance execution of low-level vision algorithms have been studied intensively. Intensive attention to other application areas has only recently developed. This bibliography makes no claim for completeness. It is intended to give a representative sample of research. BOOKS, CONFERENCE PROCEEDINGS, AND JOURNALS There are several edited books and a number of conference proceedings that collect many examples of algorithms or application-specific architectures. The following items are representative: Computer Graphics and Image Processing. (This journal includes many articles on specialized architectures for image processing.) Duff, M. J. B., ed. 1982. Languages and Architectures for Image Processing. New York: Academic Press. (Most of the early major projects for image processing systems are represented in this volume.) Matsen, F. A., and T. Tajima, eds. 1986. Supercomputers: Algorithms, Architectures and Scientific Computation. Austin, 55
OCR for page 55
56 Texas: University of Texas Press. (The chapters by Hagstrom, Kung, Leventhal et al., Reeves, and Winkler et al., are germane. The articles are mostly written by application discipline scientists.) Institute of Electrical and Electronics Engineers. 1983. IEEE Computer Society Workshop on Computer Architecture for Pattern Analysis and Image Database Management. Pasadena, Calif. October. Preston, K., and L. Uhr, eds. 1982. Multi-Computers and Image Processing. New York: Academic Press. SPIE Symposium on Architecture and Algorithms for Digital Image Processing. 1984. Proceedings. San Diego. August. Tanimoto, S. L., and A. Klinger, eds. 1980. Computer Vision: Machine Perception Through Hierarchical Computation Structures. New York: Academic Press. JOURNAL ARTICLES, CONFERENCE PROCEEDING ARTICLES, AND REPORTS This section gives a selection of articles by topic. Some effort has been made to identify continuing substantive projects. Linear Algebra Systems Armano, H., et al. 1985. (SM)2-II: A new version of the sparse matrix solving machine. Pp. 100-107 in Proceedings of the 12th Annual International Symposium on Computer Architecture. June. Chuang, H. Y. H., and G. He. 1985. A versatile systolic array for matrix computations. Pp. 315-322 in Proceedings of the 12th Annual International Symposium on Computer Architecture. June. Hoshino, T., et al. 1983. Highly parallel processor array "PAX" for wide scientific application. Pp. 95-105 in Proceedings of the 8th International Conference on Parallel Processing. August.
OCR for page 55
S7 Computations in Physics Beetem, J., et al. 1985. The GFll supercomputer. Pp. 108-113 in Proceedings of the 12th Annual International Symposium on Computer Architecture. June. (The key feature of the GFll is its configurable network switch, which allows the establishment of a set of algorithm-specific topologies for a single-instruction multiple-data processor array. Its intended application is quantum chromodynamics.) Christ, N., and A. Terrano. 1984. A very fast parallel processor. IEEE Transactions on Computers C-33:344-350. (Another quantum chromodynamics system.) Numerical Aerodynamic Simulation The National Aeronautics and Space Administration (NASA) initiated the Numerical Aerodynamic Simulation (NAS) program in 1983 to design a computational system capable of solving the Navier-Stokes equation to a degree of resolution that would permit the computer to be a major complement to wind tunnels. The report literature on the project can be obtained from NASA's Ames Research Center. WARP Project Kung, H. T. 1984. Systolic algorithms for the CMU WARP processor. Pp. 570-577 in Proceedings of the 7th International Conference on Pattern Recognition. June. (The WARP project under the direction of H.T. Kung at Carnegie-Mellon University, is developing a computation engine tWARP] that can execute the computations needed to control an autonomous load vehicle. The algorithm set includes image analysis, differential equations for mechanical control, and artificial intelligence. There is a large literature.) Symbolic Processing The three main threads of research in symbolic processing are PROLOG machines, production rule machines, and list processing (LISP) machines. The Japanese Fifth Generation Computer Project stimulated much activity in this application. We give only a slender slice of the recent literature to allow the interested reader to work his way back through references.
OCR for page 55
58 Nakagaki, R., et al. 1985. Design of a high-speed Prolog machine (HPM). Pp. 191-197 in Proceedings of the 12th Annual International Symposium on Computer Architecture. June. Stolfo, S., and D. E. Shawl 1982. DADO: A tree structured machine architecture for production systems. In Proceedings National Conference on Artificial Intelligence. Speech Recognition Speech recognition algorithms are different from vision. recent reference. We give two Anantharaman, J., and R. Bisions. 1986. A hardware accelerator for speech recognition algorithms. In Proceedings of the 13th International Symposium on Computer Architecture. June. Digital Filters Lu, H.-H., E. A. Let, and D. G. Messerschmitt. 1985. Fast recursive filtering with multiple slow processing elements. Proceedings IEEE Transactions on Circuits and Systems 32:1119-1129. Schwartz, D. A. 1985. Synchronous Multiprocessor Realizations of Shift-Invariant Flow Graphs. Ph.D. Thesis. Atlanta, Georgia: School of Electrical Engineering, Georgia Institute of Technology. June.