arXiv Analytics

Sign in

arXiv:1212.5291 [math.OC]AbstractReferencesReviewsResources

Products of random matrices and queueing system performance evaluation

N. K. Krivulin

Published 2012-12-20Version 1

We consider (max,+)-algebra products of random matrices, which arise from performance evaluation of acyclic fork-join queueing networks. A new algebraic technique to examine properties of the product and investigate its limiting behaviour is proposed based on an extension of the standard matrix (max,+)-algebra by endowing it with the ordinary matrix addition as an external operation. As an application, we derive bounds on the (max,+)-algebra maximal Lyapunov exponent which can be considered as the cycle time of the networks.

Comments: Simulation 2001: St. Petersburg Workshop on Simulation, St. Petersburg, Russia, June 18-22, 2001. ISBN 5-7997-0304-9
Journal: Proc. 4th St. Petersburg Workshop on Simulation, 2001, pp. 304-309
Categories: math.OC, cs.SY
Subjects: 15A80, 68M20, 93C65, 90B15, 37H15
Related articles:
arXiv:1212.1521 [math.OC] (Published 2012-12-07)
Bounds on mean cycle time in acyclic fork-join queueing networks
arXiv:1306.3779 [math.OC] (Published 2013-06-17, updated 2015-07-16)
Bounds on restricted isometry constants of random matrices
arXiv:2112.07005 [math.OC] (Published 2021-12-13, updated 2022-11-21)
On the gap between deterministic and probabilistic Lyapunov exponents for continuous-time linear systems