Modelling Software Reliability Growth Phenomenon In Distributed Development Environment

The present scenario of software development life-cycle has switched into a distributed environment because of the development of network technology and ever increased demand of sharing the resources to optimize the cost. In the software reliability engineering literature, few attempts have been made to model the fault testing and debugging process in a distributed development environment. One of the reasons can be attributed to the complexity involved in developing large-scale distributed systems. As a result, their testing and debugging process is influenced by many internal and external factors, all of which may not be deterministic in nature. Since reliability is the only measure of software quality, a software reliability model is needed to estimate the current reliability level and the time and resources required to achieve the objective reliability level. As the area of software fault-debugging in distributed development environment is not thoroughly investigated in current literature, even though it is estimated to have been one of the most expensive endeavor in the industry. This objective dictates developing a non-homogenous Poisson Process based testing-effort dependent software-reliability modelling approach for distributed-systems developed under imperfect-debugging environment. The resultant integrated modelling approach describes the relationship among the calendar time, the testing-effort consumption, and fault- correction/debugging process under imperfect-debugging environment. To the best of our knowledge this is the first time that this kind of integration modelling approach has been carried out for distributed systems. Fault-debugging process and testing-effort expenditures are described by a non-homogenous Poisson process and testing-effort curve functions respectively. Such a type of integrated modelling approach is very much suited also for object-oriented software development. Actual software reliability data cited in literature have been employed to demonstrate the applicability of the proposed integrated modelling approach. The results are fairly encouraging when compared with other existing approaches developed under similar environment.