The harpy machine: A data structure-oriented architecture

ABSTRACT

Efficiently storing and retrieving data has always been one of the major problems in computer technology. Until recently there has been more interest for the technology of storing and retrieving data in mass memory than for the problem of accessing data in main memory. Semiconductor technology improvements have lowered main memory cost and increased speed, therefore opening new application areas (e.g. image and speech understanding, graphics) to low cost systems. Some of the tasks in these areas are data intensive and require complex data structures. In these cases, most of the computing power is used to access data rather than to perform operations on them. The performance of such applications can be improved if the memories are tailored to the data structures they contain and to the access operations that can be performed on those data structures. The paper describes an architecture designed to explore the advantages offered by tailored, intelligent memories. The architecture has been implemented and tested on the Harpy speech understanding system. The implementation has been measured and the results are reported in the paper.

REFERENCES

• 1: J. Banerjee and D.K. Hsiao. "A Database Computer for Very Large Databases." IEEE Transactions on Computers C-28, 6 (June 1979).
• 2: R. Bisiani, The role of simulation in the development of task-oriented computer architectures, Proceedings of the Second symposium on Small systems, p.69-74, October 01-03, 1979, Dallas, Texas
• 3: R. Bisiani and H. Mauersberg. Software Development for Task-Oriented Multiprocessor Architectures. Compcon '79, 19th IEEE Computer Society International Conference, IEEE, 1979.
• 4: S.H. Lee and H. Chang. "Associative-Search Bubble Devices for Content-Addressable Memory and Array Logic." Transactions on Computers C-28, 9 (1979), 627-636.
• 5: H. Chang. Binary tree representation and manipulation in magnetic bubble memories. In preparation.
• 6: R. Greenblatt. The Lisp Machine, Artificial Intelligence Working Paper. Tech. Rept. 79, MIT, 1974.
• 7: A. K. Jones et al. "Programming Issues Raised by a Multiprocessor." Proceedings of the IEEE 66, 2 (February 1978), 229-237.
• 8: Bruce T. Lowerre, The harpy speech recognition system., 1976
• 9: Bruce Lowerre, The Harpy speech understanding system, Readings in speech recognition, Morgan Kaufmann Publishers Inc., San Francisco, CA, 1990
• 10: E. O'Neill. Pascal Microengine. Compcon'79, 19th IEEE Computer Society International Conference, IEEE, 1979.
• 11: Peter Neil Oleinick, The implementation and evaluation of parallel algorithms on c.mmp., 1979
• 12: E.A. Ozkarahan, S.A. Schuster and K.C. Smith. RAP - Associative Processor for Data Base Management. Conference Proceedings, Vol.44, AFIPS, 1975. also in NCC, June 1975, pp.379-388
• 13: B. Rosen. Architecture of a PASCAL Machine. MICRO-12, ACM, 1979.
• 14: S.Y.W. Su and J.G. Lipovski. CASSM: A Cellular System for Very Large Data Bases. International Conference on Very Large Data Bases, September, 1975.
• 15: Richard James Swan, The switching structure and addressing architecture of an extensible multiprocessor: cm*., 1978

INDEX TERMS

Primary Classification:
  I. Computing Methodologies
  I.2 ARTIFICIAL INTELLIGENCE
      I.2.7 Natural Language Processing

          Nouns: HARPY

Additional Classification:
  C. Computer Systems Organization
  C.0 GENERAL
      Subjects: System architectures

  E. Data
  E.1 DATA STRUCTURES

General Terms:
Design, Measurement, Performance