Calculating light traffic limits for sojourn times in open markovian queueing systems
Burton Simon
Abstract
Burton Simon
Abstract
Let be the m thmoment of the sojourn time distribution for some particular customer class in an open queueing system, where λ is the overall arrival rate. We derive closed form expressions, in terms of the basic system data, for the first order light traffic limit, , in the cases where the total arrival process is Poisson, a phase-type renewal process, and a superposition of independent phase-type renewal processes. For certain phase-type renewal processes, the k th order light traffic limit is zero for n≷k. In these cases we derive the kth order limit. The expressions are numerically tractable. The most difficult operation is a matrix inversion
OpenAlex reports 6 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Let be the m thmoment of the sojourn time distribution for some particular customer class in an open queueing system, where λ is the overall arrival rate. We derive closed form expressions, in terms of the basic system data, for the first order light traffic limit, , in the cases where the total arrival process is Poisson, a phase-type renewal process, and a superposition of independent phase-type renewal processes. For certain phase-type renewal processes, the k th order light traffic limit is zero for n≷k. In these cases we derive the kth order limit. The expressions are numerically tractable. The most difficult operation is a matrix inversion
Key concepts: Markovian arrival process, Superposition principle, Queueing theory, Renewal theory, Phase-type distribution, Limit (mathematics), Poisson distribution, Open system (computing)