What role can programming languages play in our continuing efforts for high‐speed computation, and what role should they play? The overwhelming majority of high‐speed computing is now done in FORTRAN on large vector machines such as the Cray 1 and the CDC Cyber 205. While the exact nature of future supercomputers is unclear, by almost all indications they will be multiprocessors. Conservative plans would link on the order of 64 processors, while more aggressive approaches, such as the “data flow” strategy, would link several thousand processors. In either case, the goal is to arrange each program to use as many processors as needed to complete execution in the shortest amount of time. To what extent can languages and their associated compilers assist us in reaching this goal?

1.
D. J. Kuck, R. H. Kuhn, D. A. Padua, B. Leasure, M. Wolfe in Proc. 8th ACM Symp. Principles of Programming Languages, Association for Computing Machinery, New York (1981), page 207.
2.
Arvind, K. P. Gostelow, W. Plouffe, An Asynchronous Programming Language and Computing Machine, Tech. Report TR114a, Dept. of Information and Computer Science, University of California, Irvine (December 1978).
3.
J. R.
McGraw
,
ACM Trans. Prog. Lang. Syst.
4
,
44
(
1982
);
W. B.
Ackerman
,
Computer
15
,
15
(
1982
).
4.
D. D.
Gajski
,
D. A.
Padua
,
D. J.
Kuck
,
R. H.
Kuhn
,
Computer
15
,
58
(
1982
).
5.
Arvind
,
K. P.
Gostelow
,
Computer
15
,
42
(
1982
).
6.
A. L.
Davis
,
R. M.
Keller
,
Computer
15
,
26
(
1982
).
7.
D. A. Turner in Proc. of the 1981 ACM Conf. on Functional Programming Languages and Computer Architectures, Association for Computing Machinery, New York (1981), page 85.
8.
A. V. Aho, J. E. Hopcroft, J. D. Ullman, The Theory of Parsing, Translation, and Compiling, Prentice Hall, Englewood Cliffs, New Jersey (1973).
This content is only available via PDF.
You do not currently have access to this content.