Paper 2, Section II, J

Applied Probability
Part II, 2016

(a) Define an M/M/M / M / \infty queue and write without proof its stationary distribution. State Burke's theorem for an M/M/M / M / \infty queue.

(b) Let XX be an M/M/M / M / \infty queue with arrival rate λ\lambda and service rate μ\mu started from the stationary distribution. For each tt, denote by A1(t)A_{1}(t) the last time before tt that a customer departed the queue and A2(t)A_{2}(t) the first time after tt that a customer departed the queue. If there is no arrival before time tt, then we set A1(t)=0A_{1}(t)=0. What is the limit as tt \rightarrow \infty of E[A2(t)A1(t)]\mathbb{E}\left[A_{2}(t)-A_{1}(t)\right] ? Explain.

(c) Consider a system of NN queues serving a finite number KK of customers in the following way: at station 1iN1 \leqslant i \leqslant N, customers are served immediately and the service times are independent exponentially distributed with parameter μi\mu_{i}; after service, each customer goes to station jj with probability pij>0p_{i j}>0. We assume here that the system is closed, i.e., jpij=1\sum_{j} p_{i j}=1 for all 1iN1 \leqslant i \leqslant N.

Let S={(n1,,nN):niN,i=1Nni=K}S=\left\{\left(n_{1}, \ldots, n_{N}\right): n_{i} \in \mathbb{N}, \sum_{i=1}^{N} n_{i}=K\right\} be the state space of the Markov chain. Write down its QQ-matrix. Also write down the QQ-matrix RR corresponding to the position in the network of one customer (that is, when K=1K=1 ). Show that there is a unique distribution (λi)1iN\left(\lambda_{i}\right)_{1 \leqslant i \leqslant N} such that λR=0\lambda R=0. Show that

π(n)=CNi=1Nλinini!,n=(n1,,nN)S\pi(n)=C_{N} \prod_{i=1}^{N} \frac{\lambda_{i}^{n_{i}}}{n_{i} !}, \quad n=\left(n_{1}, \ldots, n_{N}\right) \in S

defines an invariant measure for the chain. Are the queue lengths independent at equilibrium?