Advertisement

A preliminary evaluation of a parallel inference machine for stream parallel languages

  • Toshiaki Tarui
  • Tsutomu Maruyama
  • Hidehiko Tanaka
Parallel Architecture Aad Programming
Part of the Lecture Notes in Computer Science book series (LNCS, volume 315)

Abstract

A new architecture is proposed for a parallel inference machine for stream parallel languages. Its performance is evaluated using software simulation. In the architecture, a unit consists of a processor and a logically shared memory, and several units are connected through a network. Based on simulation results, it is confirmed that the main problem is the existence of frequent access to remote memories through the network. It is also determined that it is necessary to introduce a cluster configuration if the system is to consist of more than 64 processors.

Keywords

Shared Memory Benchmark Program Connection Time Structure Copying Inference Machine 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. [1]
    Moto-oka, T., Tanaka, H. et al, The Architecture of a Parallel Inference Engine — PIE —, FGCS '84, ICOT, 1984.Google Scholar
  2. [2]
    Ueda, K., Guarded Horn Clauses, PhD thesis, Information Engineering Course, The University of Tokyo, 1986.Google Scholar
  3. [3]
    Nilsson, M. and Tanaka, H., — FLENG Prolog — the language which turns supercomputers into parallel prolog machines, Proc. of the Logic Programming Conference '86, ICOT, 1986.Google Scholar
  4. [4]
    Hirata, K., Tanaka, H. and Moto-oka, T., A model of structure memory for PIE, Architecture Workshop In Japan '84, IPSJ, 1984 (in Japanese).Google Scholar
  5. [5]
    Sakai, S., Interconnection Networks In A Highly Parallel MIMD Computer, PhD thesis, Information Engineering Course, The University of Tokyo, 1986.Google Scholar
  6. [6]
    Yuhara, M., Koike, H., Tanaka, H. and Moto-oka, T., A Unify Processor Pilot Machine for PIE, Proc. of the Logic Programming Conference '84, ICOT, 1984.Google Scholar
  7. [7]
    Tarui, T., Maruyama, T. and Tanaka, H., Execution Method of FLENG Prolog in PIE, Proc. of 32nd Annual Convention, 5B-4, IPSJ, 1986 (in Japanese).Google Scholar
  8. [8]
    Tarui, T., Maruyama, T. and Tanaka, H., A Preliminary Evaluation of Execution Method of FLENG Prolog in PIE, Proc. of 33nd Annual Convention, 4P-11, IPSJ, 1987 (in Japanese).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • Toshiaki Tarui
    • 1
  • Tsutomu Maruyama
    • 1
  • Hidehiko Tanaka
    • 1
  1. 1.Department of Electrical EngineeringThe University of TokyoTokyoJapan

Personalised recommendations