The organization of parallel processing machines

  • A. M. Wood
Conference paper
Part of the Lecture Notes in Physics book series (LNP, volume 196)


Concurrency in computer systems can exist either as data parallelism, or instruction parallelism, or both. The Flynn scheme uses this fact along with other information to attempt a taxonomy of processors. The common usage of this scheme however, is to consider only the data and instruction parallelisms to give four mnemonics which, while being of questionable value as a classification, have become almost universally accepted.

Three operational machines have been described which fall into the parallel groupings of the Flynn mnemonics. The array processor (CLIP4) uses parallelism at the bit level to obtain very high speed processing of two-dimensional data. The pipeline machine ( Cytocomputer ) contains restricted bit-level parallelsim in the 3x3 window processors, and restricted functional parallelism in the sequence of stages. The data flow machine exhibits full functional parallelism up to the limit of the relatively few ( 15 ) processors in the system.


Graph Processing Unit Array Processor High Speed Processing Instruction Stream Functional Parallelism 
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]
    Flynn, M.J. “Some Computer Organisations and their Effectiveness” IEEE Trans. Comp. C-21 p948 1972Google Scholar
  2. [2]
    Hockney, R.W. and Jesshope, C.R. “Parallel Computers" Adam Hilger 1981Google Scholar
  3. [3]
    Danielsson, P-E. and Levialdi, S. “Computer Architectures for Pictorial Information Systems.” IEEE Computer 14 p53 1981Google Scholar
  4. [4]
    Cantoni, V. “Classification Schemes for Image Processing” Proc. NATO ASI, Cetraro, 1983Google Scholar
  5. [5]
    Duff, M.J.B. “CLIP4: A Large Scale Integrated Circuit Array Parallel Processor.” Proc. 3rd. IJCPR p728 1976Google Scholar
  6. [6]
    Wood, A.M. “The CLIP4 Array Processor” Proc. IEE Symp. on Real-time Processor Architectures...IEE London 1982Google Scholar
  7. [7]
    Batcher, K.E. “Design of a Massively Parallel Processor.” IEEE Trans. Comp. C-29 p83 1980Google Scholar
  8. [8]
    Lougheed, R.M. et al “Cytocomputers: Architectures for Parallel Image Processing.” Proc. IEEE Workshop on Picture Data Description and Management. p281 1980Google Scholar
  9. [9]
    Dennis, J.B. et al “Building Blocks for Data Flow Prototypes.” Proc. 7th Symp. Computer Arch. p1 1980Google Scholar
  10. [10]
    Arvind and Gostelow, K.P. “A Multiple Processor Dataflow Machine that Supports Generalized Procedures.” Proc. 8th Symp. Comp. Arch. p291 1981Google Scholar
  11. [11]
    Watson, I. and Gurd, J.R. “A Practical Data Flow Computer.” IEEE Computer 15 p51 1982Google Scholar

Copyright information

© Springer-Verlag 1984

Authors and Affiliations

  • A. M. Wood
    • 1
  1. 1.Dept. of PhysicsUniversity College LondonGower StreetLondon

Personalised recommendations