Metrics of Software Quality.
Abstract
The software required to test FORTRAN programs, first with random numbers and then by preselected constructed cases, was developed during this interval. Software modifications were made to permit performance measurements of the degree of coverage to include the results due to the use of constructed cases. The automation of the mix of random numbers and constructed cases was studied and limitations were found. Two new models for use during the development phase of programming were developed. These models apply to programs which grow in size during the debugging period.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 1980
- Accession Number
- ADA091385
Entities
People
- P. B. Moranda
- Z. Jelinski