Department of Computer Science Faculty of Engineering, Built Environment & IT University of Pretoria. COS122: Operating Systems

Size: px

Start display at page:

Download "Department of Computer Science Faculty of Engineering, Built Environment & IT University of Pretoria. COS122: Operating Systems"

Marshall McLaughlin
5 years ago
Views:

Department of Computer Science Faculty of Engineering, Built Environment & IT University of Pretoria COS122: Operating Systems Exam Opportunity 1 25 August 2018 Initials and Surname: Student Number:

1 Department of Computer Science Faculty of Engineering, Built Environment & IT University of Pretoria COS122: Operating Systems Exam Opportunity 1 25 August 2018 Initials and Surname: Student Number: Degree: BSc CS BSc IT BIT BIS BEng Other Examiners Mr. Christoph Stallmann, Prof. Stefan Gruner, Prof. Martin Olivier Mr. Avinash Singh, Mr. Werner Hauger, Mr. Jason Gordon Instructions 1. By attempting to write this test you declare automatically that you are in a physically and mentally healthy condition and that you consider yourself fit enough for participation. 2. Read the question paper carefully and answer all the questions which follow. 3. The examination opportunity comprises of 8 questions on 12 pages and you have 120 minutes to complete the test. 4. This is a closed book paper: you may therefore not have any study material with you. 5. You may use a non-programmable pocket calculator. Please switch off your cell phone, and keep it off for the duration of the test. 6. All examination regulations and code of conduct of the University of Pretoria are fully applicable during this test. Question: Total Marks: Score: 1

2 For the statistics of COS122, please indicate (voluntarily) whether you have a study-buddy, or whether you are studing alone without study-buddy. Please mark the answer-box accordingly: I study together with a study-buddy I study alone without a study-buddy Please also indicate (voluntarily) how many hours of preparation you have invested into this semester test: Less than 20 preparation hours Between 20 and 40 preparation hours Between 40 and 60 preparation hours More than 60 preparation hours Many thanks for your voluntary answers, which will help the Department of Computer Science with the development of the academic curriculum. 2

3 Chapter 1 - Computer System Overview 1. (a) Draw the flow chart of a read-from-memory operation with caching where data is read from (4) memory and sent to the CPU. (b) Briefly explain why information between the cache and main memory is exchanged in blocks (2) and not in bytes. Chapter 2 - Operating System Overview 2. (a) Briefly explain how an operating system works as a "shield". (2) 3

4 (b) Briefly explain how computers were used before operating systems came around. (2) (c) Briefly explain the historical connections between operating systems, software engineering, (5) and the "software crisis" of the late 1960s. (d) Draw a diagram of the multi-programming example with three processes (A, B, C) to illustrate (3) the reduction of useless waiting time when multi-programming is in place. 4

5 (e) List the three types of redundancy by means of which a system s fault tolerance can be (3) increased? Chapter 3 - Process Description and Control 3. (a) Name any four reasons for a process to terminate (or be terminated). (4) (b) Draw a six-state process state transition diagram with one suspend state to illustrate the (8) lifetime of a process in an operating system. Clearly label the states and transitions. 5

6 4. Suppose main memory contains three processes. Process A is in memory from address onwards, process B is in memory from address onwards and process C is at address onwards. Process Memory Range A B C Just before timestep 1, process A changes state to Running; at time 1 its first instruction at 2000 will be executed. Process B is at the head of the Ready queue, followed by Process C. B will begin execution at address 4000 and C will begin execution at address None of these processes have any conditional or iteration statements. The code of each process will therefore be executed from its beginning to its end. Process A terminates at instruction 2014, process B terminates at instruction 4015 and process C terminates at instruction Each instruction is executed in one (1) time unit. I/O operations also execute in one time unit, but may be blocked for a longer period of time while waiting for the device to serve the request. Assume furthermore that the dispatcher does not consume any time to execute. Hence, if say, process A executes at time 19 and the dispatcher switches execution to process B, the next instruction of process B will already be executed at time 20. Time slices have a duration of eight (8) time units. All queues are FIFO queues. The five-state process model is used. No new processes arrive during the time that this question refers to. Instructions 2002, 2013, 4004, 4014 and 6006 are I/O instructions. The I/O instructions complete during the following time units. I/O Instruction Completed during Time

7 (a) Use a program trace to determine at what time each of these I/O instructions are initiated. (5) Hint: Keep track of the ready queue to determine which of the processes will execute next. I/O Instruction Initiated at Time (b) Was the processor ever idle during this entire process? If so, during which time period? (1) Chapter 4 - Threads 5. (a) Explain the difference between the following four concepts. Be specific, vague wording like (4) "A is like B" or "A is a subset of B" will not earn you any marks. As part of your explanation, mention how these concepts are related to each other. Unit Description Program Process Thread Fibre 7

8 (b) For the two thread levels below, briefly explain the main advantage of each. The advantage (2) of one is the disadvantage of the other. Type Main Advantage User-level Threads Kernel-level Threads (c) Rank processes, user-level threads, and kernel-level threads according their the average (3) latency of operations. Latency Processes, UL threads, KL threads Slow Medium Fast (d) How can the benefits of both user-level and kernel-levels threads be achieved? (1) 6. (a) You were appointed by Google to improve the concurrency of their DeepDream engine. Google wants you to design a special version of the engine to run on AMD processors. AMD s Ryzen Threadripper can run 32 threads concurrently. Certain tasks, like loading the image files from disk, is currently executed sequentially. Calculate the relative speedup when executing DeepDream on the Threadripper if 25% of the code is sequential. Show the equation used, all steps followed, and present your final answer as a float with 3 decimal places. Also provide a recommendation to Google with a supporting argument whether or not the existing code is efficient enough to run on the 32-threaded processor. (4) 8

9 (b) What is a thread pool and what are their advantages over managing threads manually? (2) (c) How many threads are typically available in a kernel-level thread pool? Why? (2) (d) What is a zombie process? Hint: The are two characteristics applicable here. (2) (e) Name the one feature that allows virtualization in Linux? (1) (f) What is the difference between the interruptible and uninterruptible thread states in Linux? (1) (g) What is an app "window" called in Android? (1) (h) What is the difference between the visible and foreground lifetime in Android? (2) 9

10 (i) The Grand Central Dispatch in Mac allows developers to specify certain sections of the code (1) as blocks which are automatically execute in parallel. Give one word that describes this approach where a function is executed in a non-blocking fashion and can take a callback function that is executed after the block is done. Hint: Apple took this idea from the JavaScript language and it is also a common approach in RPCs. Chapter 5 - Mutual Exclusion and Synchronization 7. (a) Below are three of the six requirements that a facility providing support for mutual exclusion (3) has to meet. List the other three requirements. When no process is in a critical section, any process that requests entry to its critical section must be permitted to enter without delay. No assumptions are made about relative process speeds or number of processors. A process remains inside its critical section for a finite time only. (b) Provide the omitted code for the following struct to implement the two (counting) semaphore (5) primitives. For numbers 3 and 6, choose one of the following comments that describes the given operation in the conditional statement: Remove a process from s.queue and place the process on the ready list. Place this process in s.queue and block the process. struct semaphore int count; queuetype queue; ; void semwait(semaphore s) 1. if( 2. ) 10

11 // 3. void semsignal(semaphore s) 4. if( 5. ) // 6. Snippet/Comment Code/Explanation Snippet 1 Snippet 2 Comment 3 Snippet 4 Snippet 5 Comment 6 (c) The following illustrates the exchange instruction of mutual exclusion with hardware support. (2) Provide the missing snippets. int const n = // Number of processes int bolt; void p(int i) while( 1. ) int key = 1; do exchange( 2., 3. ) while ( 4. ); // Critical section bolt = 0; // Remainder void main() bolr = 0; parbegin (p(1), p(2),..., p(n)); 11

12 Snippet Code Snippet 1 Snippet 2 Snippet 3 Snippet 4 8. Consider the following algorithm that is intended to achieve mutual exclusion between the critical sections of Process 1 and Process 2. // Process 1 flag[0] = true; while(flag[1]) flag[0] = false; // Delay flag[0] = true; // Critical section flag[0] = false; // Process 2 flag[1] = true; while(flag[0]) flag[1] = false; // Delay flag[1] = true; // Critical section flag[1] = false; (a) Will this algorithm ever allow both processes to enter their critical sections at the same (1) time? (b) Name one reason why the algorithm is incorrect. (1) (c) Provide a sequence of events that may cause this algorithm to fail. (3) 12

Operating Systems Comprehensive Exam. Spring Student ID # 3/16/2006

Operating Systems Comprehensive Exam Spring 2006 Student ID # 3/16/2006 You must complete all of part I (60%) You must complete two of the three sections in part II (20% each) In Part I, circle or select