Relational algebra machine GRACE

  • Masaru Kitsuregawa
  • Hidehiko Tanaka
  • Tohru Moto-oka
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 147)


Most of the data base machines proposed so far which adopts a filter processor as their basic unit show poor performance for the heavy load operation such as join and projection etc., while they can process the light load operations such as selection and update for which a full scan of a file suffices. These machines which incorporates n processors takes it O(N*M/n) time to execute join of two relations whose cardinalities are N and M respectively.

GRACE adopts a novel relational algebra processing algorithm based on hash and sort, and can join in O((N + M)/n) time. GRACE exhibits high performance even in join dominant environment. In this paper, hash based relational algebra processing technique, its implementation on parallel machine, and architecture of GRACE are presented.


Data Stream Relational Algebra Memory Module Memory Bank Disk Module 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Slotnick, D.L., Logic per Track Devices, Advances in Computers, Vol. 10, J. Tou, ed., Academic Press, New York, pp. 291–296 (1970)Google Scholar
  2. 2.
    Ozkarahan, E.A., Schuster, S.A. and Smith, K.C., RAP-An Associative Processor for Data Base Management, Proc. AFIPS NCC, Vol. 45, pp. 379–387 (1975)Google Scholar
  3. 3.
    Ozkarahan, E.A., Shuster, S.A. and Sevcik, K.C., Performance Evaluation of a Relational Associative Processor, ACM Trans. Database Syst., Vol. 2, No. 2, pp. 175–195 (1977)Google Scholar
  4. 4.
    Oflazer, K. and Ozkarahan, E.A., RAP.3-A multi-micro cell architecture for the RAP database machine, Proc of the Int. Workshop on High Level Language Computer Architecture, pp. 108–119 (1980)Google Scholar
  5. 5.
    Copeland, G.P., Lipovski, G.J. and Su, S.Y.W., The Architecture of CASSM: a Cellular System for Non-numeric Processing, Proc. 1st Annu. Symp. Computer Architecture, pp. 121–128 (1973)Google Scholar
  6. 6.
    Su, S.Y.W., Nguyen, L.H., et al., The Architectural Features and Implementation Techniques of the Multicell CASSM; IEEE Trans. Comput. Vol. C-28, No. 6, pp. 430–445 (1979)Google Scholar
  7. 7.
    Su. S.Y.W., On Logic-per-Track Devices: Concept and Applications, IEEE COMPUTER, Vol. 12, No. 3, pp. 11–25 (1979)Google Scholar
  8. 8.
    Uemura, S., Yuba, T., Kokubu, A., et al., The Design and Implementation of a Magnetic-Bubble Database Machine, IFIP 80, pp. 433–438 (1980)Google Scholar
  9. 9.
    Oliver, E.J. and Berra, P.B., RELACS A Relational Associative Computer System, Proc. of the Fifth Workshop on Computer Architecture for Non-Numeric Processing, pp. 106–114 (1980)Google Scholar
  10. 10.
    Berra, P.B. and Oliver, E.J., The Role of Associative Array Processors in Data Base Machine Architecture, IEEE Computer, Vol. 12, No. 3, pp. 53–61 (1979)Google Scholar
  11. 11.
    DeWitt, D.J., DIRECT-A Multiprocessor Organization for Supporting Relational Database Management Systems, IEEE Trans. Comput., Vol. C-28, No. 6 (1979)Google Scholar
  12. 12.
    DeWitt, D.J., Query Execution in DIRECT, Proc. ACM-SIGMOD 1979, pp. 13–22 (1979)Google Scholar
  13. 13.
    Kung, H.T. and Lehman, P.L., Systolic (VLSI) Arrays for Relational Database Operations, Proc. of ACM-SIGMOD pp. 105–116 (1980)Google Scholar
  14. 14.
    Song, S.W., A Highly Concurrent Tree Machine for Database Applications, Proc. of the 1980 Int. Conf. on Parallel Processing, pp. 259–268 (1980)Google Scholar
  15. 15.
    Banerjee, J., Hsiao, D.K. and Kannan, K, DBC-A Database Computer for Very Large Databases, IEEE Trans. Comput., Vol. C-28, No. 6, pp. 414–429 (1979)Google Scholar
  16. 16.
    Menon, M.J. and Hsiao, D.K., Design and Analysis of a Relational Join Operation for VLSI, Proc. Int. Conf. on Very Large Data Bases, pp. 44–55 (1981)Google Scholar
  17. 17.
    Tanaka, Y., Nozaka, Y., et al., Pipeline Searching and Sorting Modules as Components of a Data Flow Database Computer, IFIP 80, pp. 427–432 (1980)Google Scholar
  18. 18.
    Oda, Y., Database Machine Architecture using Data Partitioning Network, IECEJ Technical Group Meeting, EC80–72 (1981) (in Japanese)Google Scholar
  19. 19.
    Babb, E, Implementing a Relational Database by Means of Specialized Hardware, ACM Trans. Database Syst., Vol. 4, No. 1, pp. 1–29 (1979)Google Scholar
  20. 20.
    McGregor, D.R., Thomson, R.H. and Dawson, W.N., High Performance Hardware for Database Systems, Systems for Large Data Bases, North-Holland, pp. 103–116 (1976)Google Scholar
  21. 21.
    Kitsuregawa, M., Suzuki, S., Tanaka, H., and Moto-oka, T., Application of Hash to a Data Base Machine, The 23rd Information Processing Society National Convention (1981) (in Japanese)Google Scholar
  22. 22.
    Kitsuregawa, M., et al., Relational Algebra Machine based on Hash and Sort, IECEJ Technical Group Meeting, EC81-35 (1981) (in Japanese)Google Scholar
  23. 23.
    Hawthorn, P., The Effect of Target Applications on the Design of Database Machines, Proc. of ACM-SIGMOD, pp. 188–197 (1981)Google Scholar
  24. 24.
    Kitsuregawa, M., Fushimi, S., Kuwabara, k., Tanaka, H., and Moto-oka, T., Organization of Pipeline Merge Sorter, IECEJ Technical Group Meeting, EC82-32 (1982) (in Japanese)Google Scholar
  25. 25.
    Todd, S., Algorithm and Hardware for a Merge Sort Using Multiple Processors, IBM J.RES.DEVELOP., Vol. 22, pp509–517 (1978)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1983

Authors and Affiliations

  • Masaru Kitsuregawa
    • 1
  • Hidehiko Tanaka
    • 1
  • Tohru Moto-oka
    • 1
  1. 1.Faculty of Engineering, Department of Information EngineeringUniversity of TokyoBunkyo-ku, TokyoJapan

Personalised recommendations