Software System Rejuvenation Modeling Based on Sequential Inspection Periods and State Multi-control Limits

This paper addresses the issue of software rejuvenation modeling. Rejuvenation strategies with sequential inspection periods and state multi-control limits are proposed here because the inspection-based approach involves the sampling of longer fixed periods of the state of system, which increases the probability of soft failure. The degradation process of the software system interferes with inspection and rejuvenation is modeled as a Markov chain. The steady-state probability density function of the system is thus derived, and a numerical solution of the function is provided. Expressions for mean unavailability time are derived during the inspection period when soft failure occurs. Finally, the steady-state availability of the system is modeled, and the solution to it is obtained using a genetic algorithm. The effectiveness of the model was verified by numerical experiments. Compared with rejuvenation strategies with fixed inspection periods, those with sequential inspection periods yielded greater steady-state availability of the software system.