Interference Analysis of Software Systems

Abstract

This paper describes a method for anticipating the occurrence of software faults in terms of a theory of dynamic strength whose functional representation has identical properties with the reliability function. Standard software complexity metrics are assessed to produce a probability distribution representing static complexity, with size as the variate, which is transformed into a complementary static strength distribution. The executing software system is also represented as a probability distribution in terms of size. With the strength and execution distributions expressed in terms of the same variate, analysis of the interference region of the distributions is performed to derive an expression for dynamic system strength. A prerequisite to the interference analysis is to properly characterize the strength distribution at various execution levels. This is accomplished by using another distribution called the return period, which is the average period of time between initializing the execution of a system and the occurrence of a failure.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 23, 2001
Accession Number
ADP011191

Entities

People

  • James O. Wilder

Organizations

  • Boeing

Tags

Communities of Interest

  • Sensors

DTIC Thesaurus Topics

  • Composite Materials
  • Data Science
  • Factor Analysis
  • Frequency
  • Gaussian Processes
  • Information Science
  • Normal Distribution
  • Probability
  • Probability Distributions
  • Reliability
  • Safety Factor
  • Software Development
  • Software Metrics
  • Software Testing
  • Standards
  • Statistical Distributions
  • Statistical Functions

Fields of Study

  • Computer science
  • Engineering

Readers

  • Applied Combinatorial Optimization and Logic Circuit Design.
  • Computational Modeling and Simulation
  • Statistical inference.