Scribd
Upload
9 views2 pages

Queueing Theory Assignment

The document outlines a series of assignments related to queueing theory, focusing on M/M/1 queues with multiple job classes, single-server queues with setup times, and networks of servers with probabilistic job routing. It includes tasks such as deriving long-run proportions of server occupancy, calculating average occupation probabilities, and finding stationary distributions for different queue configurations. The assignments require analytical derivations and references to specific sections of literature for guidance.

Uploaded by

sans.suryadevara
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
9 views2 pages

Queueing Theory Assignment

The document outlines a series of assignments related to queueing theory, focusing on M/M/1 queues with multiple job classes, single-server queues with setup times, and networks of servers with probabilistic job routing. It includes tasks such as deriving long-run proportions of server occupancy, calculating average occupation probabilities, and finding stationary distributions for different queue configurations. The assignments require analytical derivations and references to specific sections of literature for guidance.

Uploaded by

sans.suryadevara
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Assignment-1

1. Consider an M/M/1/1 queue with two classes of jobs. Class i ∈ {1, 2}


jobs arrive according to independent Poisson processes with rates λi , and
their service times are exponentially distributed with rate µi .

(a) Derive the long-run proportion of time the server is occupied by class
i jobs, for i = 1, 2.
(b) Compute both the time-average and ensemble-average occupation
probabilities for each class and verify whether they are equal or not

2. Consider a single-server queue with Poisson arrivals of rate λ. Service


times are i.i.d. with general distribution G and mean E[S]. Whenever
the system becomes empty, the server turns off immediately. Upon the
next arrival to an empty system, the server requires a random setup time
U (independent of arrivals and service times) before beginning service.
Assume E[U ] < ∞ and λE[S] < 1.
Determine the long-run fraction of time the server is:
(a) busy serving jobs,
(b) in setup,
(c) idle (powered off).
Refer to Isi mitrani section 3.2.1 for hint.
3. Consider a network of k servers, each with its own unbounded queue. Jobs
at a server are processed in First-Come, First-Served (FCFS) order. The
i-th server has an exponentially distributed service time with rate µi .
Each server may receive arrivals from both inside and outside the network.
The arrivals into the i-th server from outside the network constitute a
Poisson process with rate ri .
The routing of jobs is probabilistic. Specifically, every job that completes
service at server i will be transferred to server j with probability Pij , or
will exit the system with probability
k
X
Pi,out = 1 − Pij .
j=1

Find the stationary distribution for the above network.

1
Figure 1: An example network with only some selected Pij shown for cleanliness

4. Consider an M/M/1 queue with two classes of jobs. Class i ∈ {1, 2}


jobs arrive according to independent Poisson processes with rates λi , and
service times are exponentially distributed (with common service rate µ).
Define the state of the system at time t as the ordered sequence of job
classes currently present in the system, including the job in service. For
example, if there are three jobs in the system and the job in service is of
type 1 while the two waiting jobs are of type 2, then the state is represented
by (1, 2, 2). Derive the stationary distribution of this Markov process.

You might also like