Improved Time and Parallel Processor Bounds for Fortran-Like Loops

Author

Heuft, Richard W. ; Little, Warren D.

Author_Institution

Department of Computing Science, University of Alberta

Issue

1

fYear

1982

Firstpage

78

Lastpage

81

Abstract

Dynamic characteristics of program execution must be studied in order to calculate meaningful time and parallel processor bounds. The paper entitled "Time and Parallel Processor Bounds for Fortran-Like Loops" [1] studies only the static interaction between statements within a program to arrive at these bounds. Consequently, very loose bounds are obtained for certain programs. Improved estimates of time and processor bounds are provided for two of the example programs presented in the above-mentioned paper. These estimates are based on an expression of the relationship between operations which calculate values and operations which later use those values as operands.

Keywords

Data dependence; data flow; parallel computation; program analysis; shared variables; time and processor bounds; Design optimization; Electrons; Flip-flops; Integrated circuit interconnections; Latches; Logic circuits; Logic design; Metastasis; Tellurium; Timing; Data dependence; data flow; parallel computation; program analysis; shared variables; time and processor bounds;

fLanguage

English

Journal_Title

Computers, IEEE Transactions on

Publisher

ieee

ISSN

0018-9340

Type

jour

DOI

10.1109/TC.1982.1675890

Filename

1675890