Software Component Dependability - a Subdomain-based Theory

A proole-dependent \dependability" for software can be deened using a functional proole (a set of input-space subdomains breaking down the behavior of a component according to its speciication { these are the subdomains used in the usual \functional testing"). It is argued that measurements of dependability using the squeeze play (1) are proole dependent in general; but, (2) when the proole is deened by weighted subdomains, the squeeze-play predictions depend only on the subdomains, not on their weighting, hence this dependability is proole independent. A component developer can deene functional subdomains for the component and measure the dependability, then market the component with its dependability description. A system developer could select components with attached dependabilities, combine the components in a veriied structure, and measure { from the component dependabilities alone { a system dependability. However, the system dependability is not proole independent. Measurements for both component-and system developer can be supported by tools.

[1]  John D. Musa,et al.  Operational profiles in software-reliability engineering , 1993, IEEE Software.

[2]  John B. Goodenough,et al.  Correction to "toward a theory of test data selection" , 1975, IEEE Transactions on Software Engineering.

[3]  B. Eng,et al.  GENERATING A TEST ORACLE FROM PROGRAM DOCUMENTATION , 1995 .

[4]  G. B. Finelli,et al.  The Infeasibility of Quantifying the Reliability of Life-Critical Real-Time Software , 1993, IEEE Trans. Software Eng..

[5]  Jeffrey M. Voas,et al.  Faults on its sleeve: amplifying software reliability testing , 1993, ISSTA '93.

[6]  John D. Musa,et al.  Software reliability - measurement, prediction, application , 1987, McGraw-Hill series in software engineering and technology.

[7]  Jeffrey M. Voas,et al.  Improving the software development process using testability research , 1992, [1992] Proceedings Third International Symposium on Software Reliability Engineering.

[8]  Hélène Waeselynck,et al.  STATEMATE applied to statistical software testing , 1993, ISSTA '93.

[9]  Jeffrey M. Voas,et al.  Putting assertions in their place , 1994, Proceedings of 1994 IEEE International Symposium on Software Reliability Engineering.

[10]  Brian Marick,et al.  The craft of software testing , 1994 .