Abstract | ||
---|---|---|
In stochastic fluid models the drift at which the fluid level changes in the fluid buffer and the generator of the underlying process might depend on the discrete state of the system and on the fluid level itself. In this paper we analyse the stationary behaviour of finite buffer Markov fluid models in which the drift and the generator of the underlying continuous time Markov chain (CTMC) depends on both of these parameters. Especially, the case when the drift changes sign at a given fluid level is considered. This case requires a particular treatment, because at this fluid level probability mass might develop. When dealing with sign changes, new problems that were not addressed in previous works arises. The set of stationary equations is provided and a transformation of the unknowns is applied to obtain a solvable system description. Numerical examples introduce the behaviour of fluid systems with various discontinuities and sign changes of the drift. |
Year | DOI | Venue |
---|---|---|
2008 | 10.1016/j.peva.2007.06.027 | Perform. Eval. |
Keywords | Field | DocType |
fluid buffer,stationary distribution.,finite buffer,stochastic fluid model,drift changes sign,fluid level,stochastic ∞uid model,stationary distribution,stationary analysis,sign change,dependent bounded fluid model,fluid level probability mass,fluid system,fluid level change,markov fluid model,fluid model,continuous time markov chain | Probability mass function,Applied mathematics,Classification of discontinuities,Continuous-time Markov chain,Control theory,Markov chain,Fluid queue,Stationary distribution,Fluid models,Mathematics,Bounded function,Distributed computing | Journal |
Volume | Issue | ISSN |
65 | 3-4 | Performance Evaluation |
Citations | PageRank | References |
7 | 0.74 | 5 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
M. Gribaudo | 1 | 209 | 15.51 |
M. Telek | 2 | 228 | 19.80 |