Skip to main content
SpringerLink
Log in

An ultra-multiprocessing machine architecture for efficient parallel execution of functional languages

  • Part III Parallel Architectures And VLSI Logic
  • Conference paper
  • First Online:
Book cover Concurrency: Theory, Language, and Architecture (CONCURRENCY 1989)

Part of the book series: Lecture Notes in Computer Science ((LNCS,volume 491))

Included in the following conference series:

  • 117 Accesses

Abstract

In the execution of a functional program, a large number of function instances are dynamically created, and these created function instances are executed as fine grain concurrent processes. In order to implement massively parallel execution of such fine grain concurrent processes, ultra-multiprocessing mechanism must be designed in parallel machine architecture.

This paper proposes a machine architecture for massively parallel execution of functional programs. The machine performs parallel execution along a multithread control flow, which is called datarol. First, the datarol concept is discussed in comparison with the dataflow model. Next, a method to extract datarol program from a functional program is described through a dependency analysis. Then, a datarol machine architecture is described. The datarol processor offers a parallel execution mechanism for ultra-multiprocessing based on the continuation-based execution control mechanism. Finally, the evaluation of the datarol architecture is shown.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Dennis,J.B: A Preliminary Architecture for a Basic Data Flow Processor, Proc. 2nd Ann. Symp. on Computer Architecture, 1975, pp.126–132.

    Google Scholar 

  2. Arvind, Gostelow, K.P. and Plouffe, W.: An Asynchronous Programming Language and Computing Machine, Report TR114a, Dept. Inf. and Comp. Science, Univ. California, Irvine, 1978.

    Google Scholar 

  3. Watson,I. and Gurd,J.: A Prototype Data Flow Computer with Token Labeling, Proc. NCC, pp.623–628, 1979.

    Google Scholar 

  4. Amamiya,M., Hasegawa,R.,Nakamura,O. and Mikami,H.: A list-processing-oriented data flow machine architecture, Proc. NCC, AFIPS, 1982, pp.143–151.

    Google Scholar 

  5. Darlington,J. and Reeve,M.J.: “ALICE”: A multiprocessor Reduction Machine, Proc. Conf. Functional Programming Languages and Computer Architecture, 1982, pp.65–75.

    Google Scholar 

  6. Vegdahl, S.: A survey of proposed architecture for the execution of functional languages, IEEE Trans. on Computer, 33, 1984, pp.1050–1071.

    Google Scholar 

  7. Amamiya,M.: A New Parallel Graph Reduction Model and Its Machine Architecture, Proc. ICPP, 1987, pp.47–50.

    Google Scholar 

  8. Amamiya, M.: Data Flow Computing and Parallel Reduction Machine, Future Generations Computer Systems, Vol.4, No.1, 1988, pp.53–67.

    Article  Google Scholar 

  9. Amamiya, M.: Dataflow computing and eager and lazy evaluations, New Generation Computing, Vol.2, 1984, pp.105–129.

    Google Scholar 

  10. Amamiya,M., Takesue,M., Hasegawa,R. and Mikami,H.: Implementation and evaluation of a list-processing-oriented data flow machine, Proc. 13th Ann.Int. Conf. on Computer Architecture, 1986, pp.10–19.

    Google Scholar 

  11. Amamiya, M. Hasegawa, R. and Ono, S.: Valid: A High-Level Functional Language for Data Flow Machine, Rev. ECL, Vol.32, No.5, pp.793–802, NTT, 1984.

    Google Scholar 

  12. Sonoda,K., Ueda,T., Taniguchi,R. and Amamiya,M.: Processor Optimization and Load Control in Datarol Architecture, to appear in Proc. Joint Symposium on Parallel Processing, in Japanese, May, 1990.

    Google Scholar 

  13. Tachibana, T., Taniguchi, R. and Amamiya, M.: Compiling Method of Functional Programming Language Valid by Data Flow Analysis — Extraction of Datarol Program —, Journal of Information Processing, Vol.30, No.12, pp.1628–1638, in Japanese, 1989.

    Google Scholar 

  14. Ueda,T., Taniguchi,R. and Amamiya,M.: Datarol Processor: Its Design and Performance Evaluation, Proc. Computer System Workshop, CPSY 89-15, The Institute of Electronics, Information and Communication Engineers, in Japanese, 1989.

    Google Scholar 

  15. Nikhil,R.S. and Arvind: Can Dataflow Subsume von Neumann Computing?, Proc. Ann. Int. Symp. on Computer Architecture, pp.262–272, 1989.

    Google Scholar 

  16. Sakai,S., Yamaguchi,Y., Hiraki,K., Kodama,Y. and Yuba,T.: An Architecture of a Dataflow Single Chip Processor, Proc. Ann. Int. Symp. on Computer Architecture, pp.46–53, 1989.

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Department of Information Systems Graduate School of Engineering Sciences, Kyushu University, 6-1 Kasugakoen Kasuga-shi, 816, Fukuoka, Japan

    Makoto Amamiya & Rin-ichiro Taniguchi

Authors
  1. Makoto Amamiya
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Rin-ichiro Taniguchi
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Akinori Yonezawa Takayasu Ito

Rights and permissions

Reprints and permissions

Copyright information

© 1991 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Amamiya, M., Taniguchi, Ri. (1991). An ultra-multiprocessing machine architecture for efficient parallel execution of functional languages. In: Yonezawa, A., Ito, T. (eds) Concurrency: Theory, Language, and Architecture. CONCURRENCY 1989. Lecture Notes in Computer Science, vol 491. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-53932-8_55

Download citation

  • DOI: https://doi.org/10.1007/3-540-53932-8_55

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-53932-2

  • Online ISBN: 978-3-540-46452-5

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics

Search

Navigation