ABSTRACT
Randomization is a popular method for the transient solution of continuous-time Markov models. Its primary advantages over other methods (i.e., ODE solvers) are robustness and ease of implementation. It is however well-known that the performance of the method deteriorates with the "stiffness" of the model: the number of required steps to solve the model up to time t tends to Λt for Λt → ∞. In this paper we present a new method called regenerative randomization and apply it to the computation of two transient measures for rewarded irreducible Markov models. Regarding the number of steps required in regenerative randomization we prove that: 1) it is smaller than the number of steps required in standard randomization when the initial distribution is concentrated in a single state, 2) for Λt → ∞, it is upper bounded by a function O(log(Λt/ε)), where ε is the desired relative approximation error bound. Using dependability and performability examples we analyze the performance of the method.
- Car94.J. A. Carrasco and A. CMder6n, "Regenerative Randomization: Theory an~l applicaton exanlples," Technical Report, UPC, 1994.Google Scholar
- Cin75.E. (Tinlar, Introduction to Stochastic Processes, Prentice- Hall, 1975, pp. 371-378.Google Scholar
- Cru76.K. S. Crump, "Numerical Inversion of Laplace Transforms Using a Fourier Series Approximation," Journal of the A CM, vol. 23, pp. 89-96, 1976. Google Scholar
Digital Library
- Dij90.N. M. Dijk, "On a Simple Proof of Uniformization for Continuous and Discrete-State Continuous-Time Markov Chains," Adv. AppI. Prob., vol. 22, pp. 749-750, 1990.Google Scholar
Cross Ref
- Gra77.W. K. Grassmann, "Transient solutions in Markovian queuing systems," Comp~ft. Operatzons Res., vol. 4, pp. 47-53, 1977.Google Scholar
Cross Ref
- Gro84.D. Gross and D. R. Miller, "The randomization technique as a modelling tool and solution procedtu'e for transient Markov processes," Operations Re,., vol. 32, pp. 343-361, 1984.Google Scholar
Digital Library
- Jen53.A. Jensen, "Markoff chains as an aid in the study of Markoff processes," Skand. Akuarietidskr(ft, vol. 36, pp. 87-91, 1953.Google Scholar
- Koh82.J. Kohlas, Stoc~astlc }lfe~hod~ of Ot~~ratioT~ Re, earth, Cambridge University Press, Cambridge, 1982.Google Scholar
- Mal94.M. Malhotra, J. K. Muppala and I'~. 5. Trivedi, "'Stiffness- Tolerant Methods for Transient Analysis of Stiff Markov Chains," lllicroelectron. Rehab., vol. 34, no. 11, pp. 1825- 1841, 1994.Google Scholar
- Moo93.A. P. Moorsel and W. H. Sanders, "Adaptive Uniformization", Technical report, University of Arizona, 1993.Google Scholar
- Qur93.M. h. Qureshi and W. H. Sa,~ders, "Rewa,'d Model Solution Methods with Impulse and Rate Rewards: An Algorithm and Numerical Results," Technical Report, University of Arizona, 1993.Google Scholar
- Rei88.A. Reibman and K. S. Trivedi, "Numerical Transient Analysis of Markov Models," Comput. Operations Res., vol. 15, pp. 19-36, 1988. Google Scholar
Digital Library
- Rub93.G. Rubino and B. Sericola, "Interval Availability Distribution Computation," in Proc. 23th Int. Symp. on Fault. Tolerant Computing FTCS-~3, Toulouse, pp. 48- 55, June 1993.Google Scholar
Cross Ref
- Sou86.E. de Souza e Silva and H. R. Gall, "Calculating Cumulative Operational Time Distributions of Repairable Computer Systems," IEEE Trans. on Computers, vol. 35, pp. 322-332, 1986. Google Scholar
Digital Library
- Sou89.E. de Souza e Silva and H. R. Gall, "Calculating Availability and Performability Measures of Repairable Computer Systems using Randomization," Journal of the A CI~i, vol. 34, pp. 179-199, 1989. Google Scholar
Digital Library
Index Terms
Regenerative randomization: theory and application examples
Recommendations
Regenerative randomization: theory and application examples
Randomization is a popular method for the transient solution of continuous-time Markov models. Its primary advantages over other methods (i.e., ODE solvers) are robustness and ease of implementation. It is however well-known that the performance of the ...
Semi-Regenerative Processes with Unbounded Rewards
A semi-regenerative process SRP is combined with a reward structure such that the accumulated reward during [0, t] is the sum of a functional of the SRP and a functional of the embedded Markov renewal process MRP. For the expected discounted return a ...







Comments