The life cycle for a process consists of the following phases:

i. Use the CPU for 6 milliseconds

ii. Issue network I/O request

iii. Block for 6 milliseconds until network I/O request is done

iv. Use CPU for 2 milliseconds

v. Issue disk I/O request

vi. Block for 10 milliseconds until disk I/O is done

vii. Exit (after negligible CPU processing)

Assuming that the system has one CPU, one network I/O device, and one disk, answer the following:

a. The system is allowed to have exactly one process active at the time. This is done by starting one process, and as soon as that process exists, a new process is started, and as soon as that second process exits, a new one is started, etc. What is the steady-state utilization of the CPU, network, and disk under this scenario? What is the throughput of the system under this scenario?

b. The system is allowed to have exactly two processes active at the time (i.e., a multiprogramming level MPL=2). This is done by starting two processes, and as soon as either of these processes exists, a new process is started, etc. What is the steady-state utilization of the CPU, network, and disk under this scenario? What is the throughput of the system under this scenario?

c. Suggest a scenario that results in maximizing the steady-state utilizations of the CPU, network, and disk? What are these maxima? What MPL would result in these maximal utilizations? What is the capacity of the system under that scenario?

d. What happens if we increase the number of processes in the system beyond the MPL you obtained in part (c) above.

e. If you are allowed to double the speed of one of the three devices in the system, which one would you speed up? Doubling the speed of a device means that it would take that device half the time to complete the work requested of it. What would be the maximum utilization of the other devices under this scenario? What is the capacity of the system under this scenario?