Assignment 2 Part I Questions - This part must be done on your own. Each question is worth 20 points. a. 1. Monolithic kernel architectures tend to perform better than microkernel architectures. The main reason is because monolithic kernel loads all OS modules into one kernel process with a single address space. As a result, OS modules can directly call each other to communicate. While with the microkernel OS architecture, OS modules are separated into user mode processes and a kernel mode process which provides a minimal set of essential kernel functions to the OS user mode processes, such as inter-process communication, process scheduling, etc. With the microkernel setup, explain why frequent communication amongst the OS user mode processes might have a noticeable negative impact on the overall OS performance. b. What are some of the drawbacks of a monolithic kernel in terms of its reliability and security? 2. Processes P2 and P14 are executing. The system has two types of pending 1/0 requests. P31 (first in queue) and P55 are waiting on secondary storage (hard drive) reads. P17 is waiting for a network write to complete. P25 (first in queue) and P19 are awaiting to be scheduled for execution on the CPU. Draw a process queueing diagram for these processes. Draw an arrow showing where P17 will go once it completes. Write the state that each process is in next to the process bubble. 3. Threads vs. Processes a. Explain why communication between threads (in a same process) can be more efficient than the communication between processes. b. Explain why context switch between threads is less costly than context switch between processes. 4. Suppose you are implementing a musical composition collaboration application. It provides video conferencing and audio recording capabilities to connect composers to work on same piece of music, it automatically processes the composed musical data in the background, it frequently backups musical assets (ones not being processed) to a cloud storage, and it allows the collaborating composers to search for existing musical pieces from an online music marketplace. Composers would usually run the application on a multiple-core CPU. At run-time, the application process creates several threads, with each thread performing one of the tasks mentioned above. Which threading model (kernel-level or user-level) you would use for the above- mentioned threads? Justify your