Chapter 6 - Testing effort-dependent software reliability growth model using time lag functions under distributed environment

Abstract

Predicting software product reliability is becoming more important as the demand for zero defects grows. Computer reliability models are used for a software application to forecast reliability. In the literature, a group of models for software reliability are focused on the entire software development life cycle. On the other side, by the developer organization or by a third-party vendor large software systems are maintained. These types of systems have been in use for a long time. The organizations in charge of maintaining these programs are generally unaware of the various types of information pertaining to the implementation life cycle phases. The reliability models available for the entire production life cycle cannot be used in this situation. Predicting faults in the maintenance of such systems, on the other hand, is critical for providing trust to the system owner as well as resource planning. This chapter compares in a distributed development environment a model that depends on the “nonhomogeneous Poisson process (NHPP)” to a model with software reliability that integrates testing effort and time lag functions during the removal of a newly developed part of system. The newly developed part of the system in “the proposed model” is simplistic in nature and follows a growth curve in an exponential form. By the newly developed component it followed an S-shaped growth curve, basically it is complex in nature. During the fault removal phenomenon, we have implemented different debugging and time lag functions in a newly developed part. The model has been tested on a variety of software data sets, and the results show that the suggested model beats the existing models.