Questions 1

Questions
1. Recall the attributes set for the simulator. What is the value of mean arrival rate λ and
mean service rate μ in customers/sec ( double click on Entity server to see its
configuration)?
2. Does the queue size increase monotonically or does it reach a steady state?
What can you deduce from this outcome?
3. Why do you observe a big change in the mean delay at the beginning of simulation?
4. What is the theoretical value for mean delay Wq (queue waiting time) and how is it
calculated?
5. From the graph, read to what value mean queueing delay Wq converges to. Compare
the value with the theoretical one.
6. What is average queue size Lq in the simulations? How does it relate to the theoretical
L
q?
7. What is the server utilization? How it can be calculated theoretically?
8. Include all generated plots into the report.

Questions
1. Prove by simulation that M/M/1 has greater value for Lq than an M/M/2 with the
same ρ. Plot your results and add them to report
Remember that you need to do changes that will keep ρ same for both M/M/1 and
M/M/2. For comparison purposes, you can use results from the first question or
newly generated results for M/M/1 if needed. Note that simulation is programmed
to give error for values of arrival rate greater than 1.
2. Verify your result by calculating the theoretical values of Lq for both M/M/1
and M/M/2 cases. Present your calculations.
3. How dos the average waiting time in the queue change in M/M/2 compared to
M/M/1?
1. Find out the optimal finite buffer size where no over-flow will occur in the system.
2. Give the value in your report. Explain how you reached the value (include plots
that you used in the report). What are the factors that affect the value of queue
size?
1. What are the input parameters (arrival and service rate)
2. What are the system size parameters? C=? K=? and queue capacities and how is
each system organized
3. Run the simulation and interpret the results.
4. Which queuing strategy is more efficient?
1. What protocol is modeled with this Jackson network?
2. How does the waiting time change compared to the M/M/1 model with same input
parameters? Specify input parameters you used when running the simulation. Why
are these waiting times different?
3. Increase probability that traffic is sent to controller to 0.2. How does the waiting time
change? How does the throughput change?