Abstract
This paper discusses the design, implementation, and performance of a bus-connected multiprocessor, called Presto, for a Rete-based production system. To perform a match, which is a major phase of a production system, a Presto match scheme exploits the sub-networks that are separated by top two-input nodes and the token flow control at these nodes. Even though it possesses coarse-grained parallelism, the match scheme has high performance, providing a narrower data transfer bandwidth and fewer synchronizing opportunities. The scheme is suited for building a low-cost, compact bus-connected parallel system without shared memory or cache memory. The Presto hardware consists a maximum of 10 processing elements (PEs), each comprised of a commercial microprocessor, 4 Mbytes of local memory, and two kinds of newly developed ASIC chips which engage in memory control and bus control, respectively. Hierarchical system software is provided for interpreter program development. Performance measurement with 10 PEs demonstrates that sample programs run 5–7 times faster.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
C. L. Forgy, “Rete: A Fast Algorithm for the Many Pattern/Many Object Pattern Match Problem,” Artificial Intelligence, vol. 19, no.1, pp. 17–37, 1982.
M.I. Schor, T.P. Daly, H.S. Lee, and B.R. Tibbitts, “Advances in Rete Pattern Matching,” in Proc. AAAI-86, 3rd National Conference on Artificial Intelligence, 1986, pp. 226–232.
D.P. Miranker, “TREAT:A Better Match Algorithm for AI Production System,” in Proc. AAAI-87, 4th National Conference on Artificial Intelligence, 1987, pp. 42–47.
T. Ishida, “Optimizing Rules in Production Systems Programs,” in Proc. AAAI-88, 5th National Conference on Artificial Intelligence, 1988, pp. 699–704.
A. Gupta, “Implementing OPS5 Production Systems on DADO,” in Proc. 13th Int. Conference on Parallel Processing, 1984, pp. 83–91.
A. Gupta, C.L. Forgy, A. Newell, and R. Wedig, “Parallel Algorithm and Architectures for Rule-Based Systems,” in Proc. 13th Int. Symp. on Computer Architecture, 1986, pp. 28–37.
P.L. Butler, J.D. Allen, and D.W. Bouldin, “Parallel Architecture for OPS5,” in Proc. 15th Int. Symp. on Computer Architecture, 1988, pp. 452–457.
A. Gupta, C.L. Forgy, D. Kalp, A. Newell, and M.S. Tambe, “Parallel OPS5 on the Encore Multimax,” in Proc. 17th Int. Conference on Parallel Processing, 1988, pp. 271–280.
S. Ishigaki, I. Morihara, and K. Kushima, “KBMS-An Expert System Building Tool-,” Review of the ECL, NTT, Jpn., vol. 37, pp. 3–7, 1989.
L. Brownston, R. Farell, E. Kant, and N. Martin, “Programming Expert Systems in OPS5: An Introduction to Rule-Based Programming,” Addison-Wesley, Reading, Mass, 1985.
Motorola:, “VME Bus Specification Manual (Rev. C),” 1985.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1990 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Kikuchi, H., Yukawa, T., Matsuzawa, K., Ishikawa, T. (1990). Presto: A bus-connected multiprocessor for a rete-based production system. In: Burkhart, H. (eds) CONPAR 90 — VAPP IV. VAPP CONPAR 1990 1990. Lecture Notes in Computer Science, vol 457. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-53065-7_88
Download citation
DOI: https://doi.org/10.1007/3-540-53065-7_88
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-53065-7
Online ISBN: 978-3-540-46597-3
eBook Packages: Springer Book Archive