Software Reliability Analysis of Three Successive Generations of a Switching System

Most current approaches to software reliability evaluation are based on data collected on a single generation of products. However, many applications are developed through improvements of the existing software: to the families of products are added various generations as the need for new functionalities arises. Experimental studies dealing with the analysis of data collected on families of products are seldom reported. In this paper, we analyze the data (failure and correction reports) collected on the software of three successive generations of the Brazilian Switching System — TROPICO-R, during validation and operation. A comparative analysis of the three products is done and the main results are outlined. Emphasis is placed on the evolution of the software and the corresponding failures and corrected faults. The analysis addresses: i) the modifications introduced on system components, ii) the distribution of failures and corrected faults in the components and the functions fulfilled by the system, and iii) the evolution of the failure intensity functions.

[1]  Jean-Claude Laprie,et al.  The KAT (Knowledge-Action-Transformation) Approach to the Modeling and Evaluation of Reliability and Availability Growth , 1991, IEEE Trans. Software Eng..

[2]  Raymond Dion,et al.  Process improvement and the corporate balance sheet , 1993, IEEE Software.

[3]  Jean-Claude Laprie,et al.  Experience in software reliability: From data collection to quantitative evaluation , 1993, Proceedings of 1993 IEEE International Symposium on Software Reliability Engineering.

[4]  Mladen A. Vouk,et al.  Measuring the field quality of wide-distribution commercial software , 1992, [1992] Proceedings Third International Symposium on Software Reliability Engineering.

[5]  Michael E. Fagan Advances in software inspections , 1986, IEEE Transactions on Software Engineering.

[6]  Michael J. Flaherty,et al.  Review of Practical software metrics for project management and process improvement by Robert B. Grady, Prentice Hall, Englewood Cliffs 1992 , 1993 .

[7]  Robert B. Grady,et al.  Practical Software Metrics for Project Management and Process Improvement , 1992 .

[8]  John J. Wallace,et al.  Designing for Ultrahigh Availability: The Unix RTR Operating System , 1984, Computer.

[9]  Tze-Jie Yu,et al.  Identifying Error-Prone Software—An Empirical Study , 1985, IEEE Transactions on Software Engineering.

[10]  Karama Kanoun,et al.  A Method for Software Reliability Analysis and Prediction Application to the TROPICO-R Switching System , 1991, IEEE Trans. Software Eng..

[11]  B. S. Vianna R&D at Telebras-CPqD: the TROPICO system , 1988, IEEE International Conference on Communications, - Spanning the Universe..

[12]  Noritoshi Murakami,et al.  Software metrics using deviation value , 1987, ICSE '87.

[13]  Victor R. Basili,et al.  Software errors and complexity: an empirical investigation0 , 1984, CACM.

[14]  Hermann Kopetz,et al.  Dependability: Basic Concepts and Terminology , 1992 .

[15]  Jean-Claude Laprie For a product-in-a-process approach to software reliability evaluation , 1992, [1992] Proceedings Third International Symposium on Software Reliability Engineering.

[16]  Karama Kanoun,et al.  Software Failure Data Analysis of two Successive Generations of a Switching System , 1993, SAFECOMP.

[17]  Bev Littlewood,et al.  Measurement for software control and assurance , 1989 .