Role of System Simulation in Analyzing Software Reliability

Increasing global competition and high development costs have intensified the pressure to quantify software quality and to measure and control the level of quality delivered. As our global society becomes more dependent on information in the production of goods and services, the pressure for higher quality, lower cost, and faster delivery for software products is increasing. Software reliability is the most important and most measurable aspect of software quality and it is highly customer oriented. According to ANSI, it is defined as the probability of failure-free software operation for a specified period of time in a specified environment. It is also the measure of how well the program functions to meet its operational requirements. Numerous software reliability growth models (SRGMs) have been developed during the last three decades providing useful information about how to improve reliability (Musa et al.(1987), Xie(1991), Lyu(1996)).These software reliability models attempt to assess expected reliability or future operability using observed failure data and statistical inference techniques. They are restricted in their life cycle scope and adaptability to general use for a number of reasons. Simulation methods are generally employed in reliability studies when deterministic methods are incapable of modelling strong dependencies between failures. Simulation is the process of constructing a model of a system which contains a problem and conducting experiments with the model on a computer for a specific purpose of experimentation to solve the problem. Reliability modelling ultimately requires good data. But software projects do not always collect data sets that are comprehensive, complete, or consistent enough for effective modelling research or model application which normally is not made available. Further, data required for software reliability modelling in general seem to be even more difficult to collect than other software engineering data. Since good data sets are so scarce, one purpose of simulation is to supply carefully controlled, homogeneous data or software artifacts having known characteristics for use in evaluating the various assumptions upon which existing reliability models have been built. Since actual software artifacts(such as faults in computer programs) and processes(such as failure and fault removal) often violate the assumptions of analytic software reliability models, simulation can perhaps provide a better understanding of such assumptions and may even lead to a better explanation of why some analytic models work well in spite of such violations. The aim of the work is to explore the goals and limitations of using simulation in analysing software reliability. Efforts are made to review the work that has been previously done in all the dimensions of the concerned research area. Furthermore emphasis is given to understand the simulation as a tool in software reliability assessment and to make a comparative study of analyzing software reliability with simulation and without simulation