Abstract
A mean-field extension of the queueing system (GI/GI/1) is considered. The process is constructed as a Markov solution of a martingale problem. Uniqueness in distribution is also established under a slightly different set of assumptions on intensities in comparison with those required for existence.
Similar content being viewed by others
References
Baccelli, F., Rybko, A.N., Shlosman, S.B.: Queueing networks with mobile servers: the mean-field approach. Probl. Inf. Transm. 52(2), 178–199 (2016)
Billingsley, P.: Convergence of Probability Measures. Wiley, New York (1999)
Borovkov, K.A.: Propagation of chaos for queueing networks. Theory Probab. Appl. 42(3), 385–394 (1998)
Gnedenko, B.V., Kovalenko, I.N.: Introduction to Queueing Theory, 2nd edn. Birkhäuser, Boston (1989)
Gnedenko, B.V., Belyaev, Yu.K., Solovyev, A.D.: Mathematical Methods in Reliability Theory. Academic Press, New York (1969)
Davis, M.H.A.: Piecewise-deterministic Markov processes: a general class of non-diffusion stochastic models. J. R. Stat. Soc. Ser. B (Methodological) 46(3), 353–388 (1984)
Dawson, D.A., Tang, J.Q., Zhao, Y.Q.: Balancing queues by mean field interaction. Queueing Syst. 49(3), 335–361 (2005)
Ethier, S.N., Kurtz, T.G.: Markov processes: characterisation and convergence. Wiley, Hoboken (2005)
Kac, M.: Foundations of kinetic theory. In: Proceedings of the Third Berkeley Symposium on Mathematical Statistics and Probability, vol. 3, pp. 171–197. University of California Press, California (1956)
Karpelevich, F.I., Rybko, A.N.: Asymptotic behavior of the thermodynamical limit for symmetric closed queueing networks. Probl. Inf. Transm. 36(2), 154–179 (2000)
Krylov, N.V.: On the selection of a Markov process from a system of processes and the construction of quasi-diffusion processes. Math. USSR—Izvestija 7(3), 691–709 (1973)
Liptser, R.Sh., Shiryaev, A.N.: Theory of Martingales. Kluwer Academic Publishers Group, Dordrecht (1989)
McKean, H.P.: A class of Markov processes associated with nonlinear parabolic equations. Proc. Natl. Acad. Sci. USA 56(6), 1907–1911 (1966)
Mishura, Yu.S., Veretennikov, A.Yu.: Existence and uniqueness theorems for solutions of McKean–Vlasov stochastic equations, arXiv:1603.02212
Rybko, A., Shlosman, S., Vladimirov, A.: Poisson hypothesis for open networks at low load. Mosc. Math. J. 17(1), 145–160 (2017)
Skorokhod, A.V.: Studies in the Theory of Random Processes. Addison-Wesley, Reading (1965)
Veretennikov, A.Yu., Zverkina, G.A.: Simple proof of Dynkin’s formula for single-server systems and polynomial convergence rates. Markov Process. Relat. Fields 20, 479–504 (2014)
Vlasov, A.A.: On Vibration properties of electron gas. J. Exp. Theory Phys. 8(3), 291–318 (1938). (in Russian)
Acknowledgements
The techniques used in this paper were stimulated by the methods developed in a long-term joint work on formally quite different McKean–Vlasov SDE equations with Yu. Mishura, as well as in fruitful discussions of the author on the same subject with D. Šiska, and L. Szpruch. S. Pirogov, A. Rybko, and G. Zverkina helped to find some (quite a few) technicalities to be corrected in the earlier versions of the text. The author is sincerely thankful to all these colleagues, as well as to the anonymous referee. The deepest gratitude is to Professor Alexander Dmitrievich Solovyev (06.09.1927—06.04.2001) who was the author’s supervisor at BSc and MSc programmes at University.
Author information
Authors and Affiliations
Corresponding author
Additional information
In memory and to the 90th anniversary of Alexander Dmitrievich Solovyev.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
This study has been funded by the Russian Academic Excellence Project ‘5-100’ and by the RFBR Grant 17-01-00633_a.
Appendix
Appendix
The following celebrated Lemma is stated for the convenience of the reader [2, Theorem 25.6].
Lemma 1
(Skorokhod [16, Ch.1, §6]) Let \(\xi ^n_t\) (\(t\ge 0\), \(n=0,1,\ldots \)) be some d-dimensional stochastic processes defined on some probability space, and let the following hold true for any \(T>0\), \(\epsilon > 0\):
Then, there exists a subsequence \(n'\rightarrow \infty \) and a new probability space can be constructed with processes \({\tilde{\xi }}^{n'}_t, \, t\ge 0\), and \({\tilde{\xi }}^{}_t, \, t\ge 0\), such that all finite-dimensional distributions of \({\tilde{\xi }}^{n'}_{\cdot }\) coincide with those of \(\xi ^{n'}_{\cdot }\) and such that, for any \(\epsilon >0\) and all \(t\ge 0\),
Rights and permissions
About this article
Cite this article
Veretennikov, A.Y. On mean-field (GI/GI/1) queueing model: existence and uniqueness. Queueing Syst 94, 243–255 (2020). https://doi.org/10.1007/s11134-019-09626-x
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11134-019-09626-x