Modeling Perfect and Minimal Rejuvenation for Client Server Systems with Heterogeneous Load

In the present study a client-server system is considered, which experiences resource degradation as clients' requests increase. Software rejuvenation is performed in order to counteract resource exhaustion. Two different levels of rejuvenation actions are implemented, perfect, and minimal. Moreover the concept of a failed rejuvenation is introduced to model the fact that rejuvenation due to some circumstances cannot be accomplished. As the load of such a system varies from hour to hour but reveals a cyclic behavior from day to day, different rejuvenation policies for each period of the day are proposed using a cyclic non-homogeneous Markov (CNHM) model. As a measure of performance, the steady-state expected downtime cost is considered. Additionally, to set off the need of CNHM modeling, the system is also modeled by a Homogeneous Markov Chain (HMC) and the performance results are compared.

[1]  George A. Gravvanis,et al.  On the optimization of free resources using non-homogeneous Markov chain software rejuvenation model , 2007, Reliab. Eng. Syst. Saf..

[2]  Agapios N. Platis.  A generalized formulation for the performability indicator , 2006, Comput. Math. Appl..

[3]  Hans-Juergen Boehm,et al.  Garbage collection in an uncooperative environment , 1988, Softw. Pract. Exp..

[4]  Yennun Huang,et al.  Software rejuvenation: analysis, module and applications , 1995, Twenty-Fifth International Symposium on Fault-Tolerant Computing. Digest of Papers.

[5]  George Candea,et al.  Recovery-oriented computing: building multitier dependability , 2004, Computer.

[6]  Tadashi Dohi,et al.  Estimating Periodic Software Rejuvenation Schedules under Discrete-Time Operation Circumstance , 2008, IEICE Trans. Inf. Syst..

[7]  Kishor S. Trivedi,et al.  An approach for estimation of software aging in a Web server , 2002, Proceedings International Symposium on Empirical Software Engineering.

[8]  Nikolaos Limnios,et al.  Performability of electric-power systems modeled by non-homogeneous Markov chains , 1996, IEEE Trans. Reliab..

[9]  Tadashi Dohi,et al.  A Faster Estimation Algorithm for Periodic Preventive Rejuvenation Schedule Maximizing System Availability , 2007, ISAS.

[10]  Kishor S. Trivedi,et al.  Performability Analysis: Measures, an Algorithm, and a Case Study , 1988, IEEE Trans. Computers.

[11]  A.N. Platis,et al.  Asymptotic availability of systems modeled by cyclic non-homogeneous Markov chains [substation reliability] , 1997, Annual Reliability and Maintainability Symposium.

[12]  Elaine J. Weyuker,et al.  Monitoring Smoothly Degrading Systems for Increased Dependability , 2004, Empirical Software Engineering.

[13]  Jordi Torres,et al.  Using Virtualization to Improve Software Rejuvenation , 2007, IEEE Transactions on Computers.

[14]  Dong Seong Kim,et al.  Software Rejuvenation and Reconfiguration for Enhancing Survivability of Sensor Networks , 2008, 22nd International Conference on Advanced Information Networking and Applications - Workshops (aina workshops 2008).

[15]  Wei Xie,et al.  Analysis of a two-level software rejuvenation policy , 2005, Reliab. Eng. Syst. Saf..

[16]  Kishor S. Trivedi.  Probability and Statistics with Reliability, Queuing, and Computer Science Applications , 1984 .

[17]  Agapios N. Platis,et al.  Semi-Markov Availability Modeling of a Redundant System with Partial and Full Rejuvenation Actions , 2008, 2008 Third International Conference on Dependability of Computer Systems DepCoS-RELCOMEX.

[18]  Nikolaos Limnios,et al.  Dependability analysis of systems modeled by non-homogeneous Markov chains , 1998 .

[19]  Kishor S. Trivedi,et al.  Analysis of inspection-based preventive maintenance in operational software systems , 2002, 21st IEEE Symposium on Reliable Distributed Systems, 2002. Proceedings..