| ¡¡ | Chinese Journal of Computers Full Text |
| Title | Modeling and Analysis of Software Rejuvenation Based on Non-Markovian Stochastic Petri Nets |
| Authors | MENG Hai-Ning QI Yong HOU Di |
| Address | (School of Electronics and Information Engineering, Xi¡äan Jiaotong University, Xi¡äan 710049) |
| Year | 2007 |
| Issue | No.12(2212¡ª2217) |
| Abstract & Background | Abstract Software aging is an important potential factor that affects the software reliability. As a proactive and preventive software fault tolerant technique, software rejuvenation is a main method for counteracting software aging. Almost all developed software rejuvenation models based on stochastic Petri Nets assume that all the firing times submit to exponential distributions. Aiming at the firing times submit to determined or general distributions, a software rejuvenation modeling method is proposed using Non-Markovian Stochastic Petri Nets. In addition, model is solved for both steady and transient state via Markov regenerative theory. The numeric experiment results show that selecting optimal software rejuvenation schedule can improve systematic availability and reduce downtime cost. keywords software aging; software rejuvenation; software reliability; non-Markovian stochastic Petri nets; Markov regenerative theory background Recent studies have reported the phenomenon of software aging in which the state of system performance degrades with time. In order to enhance software reliability and prevent system degradation or crash, such a preventive maintenance technique called software rejuvenation was introduced which provides the theory basis and the method for fault-recover and self-healing. Under the National "863" Foundation Project (2001AA113040), the National Natural Foundation of China (60473098) and IBM Joint Research, we developed the technical achievement and conduct the research on software rejuvenation. |