MACH: Much Faster Associative Machine

  • Ryohei Nakano
  • Minoru Kiyama
Part of the The Kluwer International Series in Engineering and Computer Science book series (SECS, volume 43)

Abstract

This paper proposes a new database machine architecture called MACH (Much Faster Associative Machine), the goal of which is to improve relational performance by two orders. This architecture is aimed mainly at the knowledge processing field, where such high performance is required. The design principles are first presented along with an overview of MACH architecture. After which, the main characteristics of MACH architecture are described in detail, including its memory resident database, fixed-length encoding, sophisticated data storing, and hash-based algorithms for main relational algebra operations. Experiment results gained from encoding databases in practical use are also shown. Tests conducted in the initial implementation of MACH1 showed that its performance exceeds any disk-based machine or system by more than one order.

Keywords

Compaction Volatility Suffix Hine 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Agrawal, R. and DeWitt, D.J., “Whither Hundreds of Processors in a Database Machine?,” Int’l Workshop on High-Level Architecture, Los Angeles,Cal.,1984.Google Scholar
  2. [2]
    Fushimi, S., Kitsuregawa, M., and Tanaka, H., “An Overview of the System Software of a Parallel Relational Database Machine GRACE,” Proc.12th VLDB, Kyoto, Japan, May 1986.Google Scholar
  3. [3]
    Bitton, D., DeWitt, D.J., and Turbyfill, C., “Benchmarking Database Systems: A Systematic Approach,” CS Tech. Rep. #526 Univ. of Wisconsin-Madison, Dec. 1983.Google Scholar
  4. [4]
    Yokota, K., “Deductive Approach for Unnormalized Model,” SIG Notes IPS Japan, 87-DB-58, Mar. 1987 (in Japanese).Google Scholar
  5. [5]
    Nakano, R., and Kiyama, M., “Frame Calculus,” Sig Notes IPS Japan, 87-AI-50, 1987 (in Japanese).Google Scholar
  6. [6]
    Leland, M.D.P., and Roome W.D., “The Silicon Database Machine,” Proc. 4th IWDM, Grand Bahama Island ,March 1985.Google Scholar
  7. [7]
    DeWitt, D.J., et al., “Imple mentation Tecniques for Main Memory Database Systems,” SOGMOD’84 Boston,MA, pp.1–17,June 1984.Google Scholar
  8. [8]
    Codd, E.F., “Relational Completeness of Data Base Sublanguages,” in Data Base Systems, Courant Computer Symposium 6,Prentice Hall, 1972.Google Scholar
  9. [9]
    ISO/TC97/SC21, “Database Language SQL,” ISO/TC97/SC21/WG 5–15, 1985.Google Scholar
  10. [10]
    Nakano, R., and Saito, K., “Rule-Based Reduction From Relational Calculus to Succinct Relational Algebraic Expression,” Sig Notes IPS Japan,86-DB-54, 1986 (in Japanese).Google Scholar
  11. [11]
    Nakano, R., and Saito, K., “Reduction of Aggregate Functions in Relational Calculus to Optimal Algebraic Expressions,” Sig Notes IPS Japan,87-DB-57, 1987 (in Japanese).Google Scholar
  12. [12]
    Jaeschke, G. and Schek, H.-J., “Remarks on Algebra of Non First Normal Form Relations,” Proc. ACM Symposium on Principles of Database Systems, Calif., Mar. 1982.Google Scholar
  13. [13]
    Schek, H.-J. and Pistor, P., “Data Structures for an Integrated Data Base Management and Information Retrieval System,” Proc. 8th VLDB, Mexico City,Mexico, Sep. 1982.Google Scholar
  14. [14]
    Ogura, T., et al., “A 4-Kbit Associative Memory LSI,” IEEE J. Solid State Circuits, SC-20,6,pp.1277–1282 ,1985.CrossRefGoogle Scholar
  15. [15]
    Tanaka Y., “A Data-Stream Database Machine with Large Capacity,” in Advanced Database Machine Architecture, D.K. Hsiao (ed.), Prentice-Hall,pp.168–202 ,1983.Google Scholar
  16. [16]
    Boyer, R.S. and Moore, J.S., “A Fast String Searching Algorithm,” Comm.ACM, 20, 10, pp.762–772, 1977.MATHCrossRefGoogle Scholar
  17. [17]
    Nakano, R., and Kiyama, M., “Experiment of Encoding Databases,” Sig Notes IPS Japan, 87-FI-4, 1987 (in Japanese).Google Scholar
  18. [18]
    Itano K., et al., “A Pipelined String Search Algorithm Based on an Associative Memory,” Trans.IPS Japan,26,6,pp.1152–1155 ,1985 (in Japanese)Google Scholar
  19. [19]
    Larson, P.-A, “Dynamic Hashing,” BIT,18,2,pp.184–201 ,1978.MathSciNetMATHCrossRefGoogle Scholar
  20. [20]
    Litwin, W., “LINEAR HASHING:A New Tool for File and Table Addressing,” Proc. 6th VLDB, pp.212–223, 1980.Google Scholar
  21. [21]
    Knuth, D.E., “The Art of Computer Programming,Vol.2:Semi-numerical algorithms,” Addison-Wesley,Reading, Mass. ,1973.Google Scholar
  22. [22]
    Murakami, K., et al., “A Relational Database Machine: First Step to Knowledge Base Machine,” Proc. 10th Symposium on Computer Architecture,June 1983.Google Scholar
  23. [23]
    Lehman, T.J., and Carey, M.J., “Query Processing in Main Memory Database Systems,” SIGMOD’86 Washington,DC, pp.239–250, March 1986.Google Scholar
  24. [24]
    Nakano, R., and Kiyama, M., “Full Associative Processing for Relational Operations,” 34th Annual Convention IPS Japan,3C-1 ,March 1987 (in Japanese).Google Scholar

Copyright information

© Kluwer Academic Publishers, Boston 1988

Authors and Affiliations

  • Ryohei Nakano
    • 1
  • Minoru Kiyama
    • 1
  1. 1.NTT Communications and Information Processing LaboratoriesYokosuka, KanagawaJapan

Personalised recommendations