Logarithmic asymptotics for a single-server processing distinguishable sources
- 40 Downloads
We consider a single-server first-in-first-out queue fed by a finite number of distinct sources of jobs. For a large class of short-range dependent and light-tailed distributed job processes, using functional large deviation techniques we prove a large deviation principle and logarithmic asymptotics for the joint waiting time and queue lengths distribution. We identify the paths that are most likely to lead to the rare events of large waiting times and long queue lengths. A number of examples are presented to illustrate salient features of the results.
KeywordsFunctional large deviations Single server FIFO Waiting time Queue length
Mathematics Subject Classification (2000)Primary 60K25 Secondary 60F10 90B05
Unable to display preview. Download preview PDF.
- Dembo A and Zeitouni O (1998). Large deviation techniques and applications. Springer, Berlin Google Scholar
- Ganesh A, O’Connell N and Wischik D (2004). Big queues. Lecture Notes in Mathematics, vol 1838. Springer, Berlin Google Scholar
- Glynn PW, Whitt W (1994) Logarithmic asymptotics for steady-state tail probabilities in a single-server queue. J Appl Probab 31A:131–156, Studies in applied probabilityGoogle Scholar
- Mogul′skiĭ AA (1974). Large deviations in the space C(0, 1) for sums that are defined on a finite Markov chain. Sibirsk. Mat. Ž. 15(61–75): 237 Google Scholar
- Rockafellar TR (1970). Convex analysis, Princeton Mathematical Series, No. 28. Princeton University Press, Princeton Google Scholar