Communication in computation

  • Robert W. Keyes


The need for communication is an essential factor shaping the physical nature of a computing system. A computing system functions by bringing a large number of separate elements to bear on a common problem. Coordinated operation of the elements requires a large amount of communication among them through many long wires. Some implications of communication for the physical characteristics of electrical computers are as follows. Variability in manufacturing and of the environment within a system causes the elements of a system to differ from one another in their responses to electrical signals. Signals must be large enough to be interpretable by any element of the system. Therefore, the need for rapid reliable communication implies that signals that are large in some sense are required. The combination of large signals and many wires implies high-power dissipation. Fitting the pattern of interconnection into minimum space is a very complex problem which is one of the most severe limitations on the utilization of integrated electronics in computation. Experience has led to the establishment of certain quantitative generalizations concerning the communication requirements in large systems. Similarity between these and neuronal communication is noted.


Quantum Field Theory Computing System Physical Characteristic Complex Problem System Function 
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. Beall, R. J. (1974). “Packaging for a supercomputer,” paper 18/31974 INTERCON Technical Papers. IEEE. New York.Google Scholar
  2. Blumberg, R. J., and Brenner, S. (1979). “A 1500 gate random logic LSI masterslice”IEEE Journal of Solid State Circuits,14, 818–822.Google Scholar
  3. Bok, S. T. (1959). Elsevier, Amsterdam, pp. 185–232.Histonomy of the Cerebral Cortex,Google Scholar
  4. Chiba, T. (1978). “Impact of the LSI on high-speed computer packaging,”IEEE Transactions on Computers,27, 319–325.Google Scholar
  5. Dansky, A. H. (1980). “Bipolar circuit design for VLSI gate arrays,” pp. 674–677.Proceedings ICCC 80. IEEE, New York.Google Scholar
  6. Donath, W. E. (1968). “Statistical properties of the placement of a graph,”SIAM Journal of Applied Mathematics,16, 376–387.Google Scholar
  7. Donath, W. E. (1979). “Placement and average interconnection lengths of computer logic,”IEEE TRansactions of Circuits and Systems,26, 722–277.Google Scholar
  8. Edwards, N. P. (1978). Unpublished.Google Scholar
  9. El Gamal, A. A. (1981). “Two-dimensional stochastic model for interconnection in integrated circuits,”IEEE Transactions on Circuits and Systems,28, 127–135.Google Scholar
  10. Heller, W. R., Mikhail, W. F., and Donath, W. E. (1977). “Prediction of wiring space requirements for LSI,” pp. 32–42.Proceedings 14th Design AutomationConference, June 20–22, 1977, New Orleans.Google Scholar
  11. Jerison, H. J. (1973).Evolution of the Brain and Intelligence, pp. 68 and 69. Academic Press, New York.Google Scholar
  12. Landman, B. S., and Russo, R. L. (1971). “On a pin versus block relationship for partitions of logic graphs,”IEEE Transactions on Computers,20, 1469–1479.Google Scholar
  13. Sutherland, I., and Oestreicher, D. (1973). “How big should a printed circuit board be?”IEEE Transactions on Computers,22, 537–542.Google Scholar
  14. Werbizky, G. G., Winkler P. E., and Haining, F. W. (1980). “Packaging technology for the IBM 4300 processors,”1980 Spring COMPCON, pp. 304–311. IEEE, New York.Google Scholar
  15. Wilson, E. A. (1977). “True liquid cooling of computers”,1977 National Computer Conference Proceedings, Montvale, New Jersey, pp. 341–348, AFIPS Press.Google Scholar
  16. Wu, L. C. (1978). “VLSI and mainframe computers,”Spring 78 Compcon Digest, pp. 26–29. IEEE, New York.Google Scholar

Copyright information

© Plenum Publishing Corporation 1982

Authors and Affiliations

  • Robert W. Keyes
    • 1
  1. 1.IBM T. J. Watson Research CenterYorktown Heights

Personalised recommendations