Title :
Good System Structure Features: Their Complexity and Execution Time Cost
Author :
Stankovic, John A.
Author_Institution :
Departmnent of Electrical and Computer Engineering, University of Massachusetts
fDate :
7/1/1982 12:00:00 AM
Abstract :
This paper describes a multistep technique that can be applied to improve system structure and to improve performance when necessary. The technique begins with the analysis of system structure via the structured design guidelines of coupling and cohesion. Next, manual system structure improvement transformations are applied. The effect of the transformations on execution time is then determined. Finally, vertical migration is used on the restructured system to improve its performance. Using the results of this paper, system programmers can identify specific cases where both good system structure and good performance are attainable, and others where tradeoffs must be made. The technique is most applicable during the maintenance phase of the software life cycle.
Keywords :
Execution time performance; hierarchical model; large systems; maintenance; software complexity; software methodology; structured design; transformations; vertical migration; Application software; Costs; Design methodology; Guidelines; Hardware; Programming profession; Software maintenance; Software performance; Software systems; Virtual manufacturing; Execution time performance; hierarchical model; large systems; maintenance; software complexity; software methodology; structured design; transformations; vertical migration;
Journal_Title :
Software Engineering, IEEE Transactions on
DOI :
10.1109/TSE.1982.235425