Sample Exam Questions
CSE2302 Operating Systems, Semester 2, 2005

The examination paper is out of 267 marks has two sections. Only integral marks are given.

Part A contains 33 multi-set questions, where more than one alternative can be correct. Each question is worth 3 marks, giving a possible 99 marks in all. You answer by circling what you think are the correct answers. To get full marks all and only correct alternatives must be circled.

Part B contains 8 questions with written answers worth a possible 168 marks. Answers are written in within the examination paper.

The following instructions to candidates are taken from the examination paper:

Instructions to Candidates

  1. Write your student ID and desk number in the section above.
  2. Answer all questions in the space provided.
    Reverse side of question paper may be used for rough working but will not be marked.
  3. There are two sections: A and B.
    Section A is worth 99 marks and
    Section B is worth 168 marks.
  4. Total marks for this examination are 267.
  5. Calculators are NOT allowed.
  6. This paper is NOT to be removed from the examination room.

Sample Questions for Part A

Sample answers to these quesions can be found here.

1. The Windows and UNIX shell programs
a. are not processes at all
b. provide the user with system administrator privileges
c. are executed by users and provide user privileges
d. are the programs which provide a command line interface for all users
e. none of the above

2. The operating system
a. is important because it provides an interface between the user and the computer hardware
b. is difficult to define
c. is defined as the composition of process manager, memory manager and I/O manager
d. causes the termination of a processe which has a runtime SEGFAULT error
e. is/does none of the above

3. Which of the following can be described as a resource managed by the operating system:
a. the number of keys on the keyboard
b. the amount of memory for a process
c. the number of instructions which can executed by the CPU per second
d. the file-system
e. the file-system interface

4. Which OS strategy assures users that a service will be delivered at a pre-specified time?
a. batch processing
b. timesharing
c. interactive data processing
d. real time scheduling
e. none of the above

5. Which if the following statements is true about multiprogramming?
a. it makes threads obsolete
b. it can be used to save memory
c. it increases the systems throughput
d. it increases the efficiency of the CPU
e. it increases the efficiciency of memory

6. Turn-around time is
a. time taken to load a program into memory
b. time from arrival of the process to termination of the process
c. time from arrival of the process to first access to the CPU
d. is not a usefull concept
e. is measured in microseconds

7. On a single CPU multiprogrammed system,
a. we can have several processes in the run state at once
b. there is a queue of running processes
c. there is a queue of ready processes
d. only one process can be blocked for I/O at any time
e. there is a queue for processes that are blocked for I/O

8. Processes that are sleeping
a. are in the ready queue
b. can wake themselves
c. must be woken by an external event
d. are executing a spin-lock
e. can be waiting for I/O

9. The short-term scheduler
a. schedules suspended processes
b. is used to schedule processes between the ready ready state and the running state
c. allocate process IDs
d. blocks processes for I/O operations
e. none of the above

10. A directory is
a. a file that contains only file names
b. a data structure that allows indexed access to file contents
c. a data structure that provides indexed access to file names
d. able provide for a hierarchical organisation of filenames
e. only useful when using UNIX

11. A file is
a. the underlying storage mechanism for a database management system
b. a named, linear sequence of bytes
c. is used to contain persistent data
d. is used to protect persistent data
e. none of the above

12. Which of the following are distinct components of a process?
a. the run-time image on the disk
b. code and data only
c. code, data and stack
d. list of resources
e. the process control block

13. The CPU can execute code
a. from the disk drive
b. only when it resides in memory
c. only after the process control block has been generated and the code is in memory
d. only when the execute permission is enabled

14. When a process is terminated
a. it removes itself from memory
b. it is dequeued from the ready queue
c. it is removed from memory and put in swap space on a disk
d. it stops executing but remains in memory

15. A CPU protection mechanism
a. is not really a problem for the OS
b. is critical for the safe and correct execution of the OS
c. protects users from the OS
d. protects the OS from inadvertant damage cause by user processes

16. Segmented virtual memory
a. divides memory into fixed sized frames
b. divides code into fixed sized segments
c. allows for variable sized segments
d. reduces the amount of internal fragmentation

17. A user mode program can invoke OS services by:
a. having system administrator permission
b. message passing
c. executing a trap instruction
d. executing a system call
e. none of the above

18. Consider the relationships amongst the major OS components:
a. the process manager is inseparable from the file manager
b. the device manager relies on the memory manager
c. the UNIX file manager uses the device manager
d. none of the above

19. In a paged (fixed size) memory partitioning scheme what causes fragmentation?
a. very large processes
b. large number of very small processes
c. a mismatch between memory requests and the size of memory partitions
d. processes which are one byte larger than an integral number of pages

20. The DMA controller is used to
a. increase the efficiency of the CPU
b. transfer data from peripheral to peripheral
c. cannot be used directly by the OS
d. transfer data from peripherals to memory
e. transfer code from memory to swap space

21. The translation look-aside buffer
a. provides a means of managing access to data on the disk
b. is a cache used to hold recently referenced page table entries
c. reduces the frequency of access to main memory to retrieve page table entries
d. is implemented in the memory management unit of the CPU
e. none of the above

22. A page fault
a. is an error in the code within a page in main memory
b. is used to begin the execution of code that brings a page into memory
c. will be the cause of thrashing when memory is over allocated
d. is usually going to lead to a process switch

23. A context switch is
a. the procedure of saving the execution state of a process and restoring the execution state of another process
b. to turn the computer off
c. the procedure of moving a program out of memory
d. caused by a periodic timer interrupt
e. none of the above

24. A CPU with interrupts usually includes instructions to enable and disable interrups
a. so that we can synchronize with the power cycle
b. to prevent race conditions in a multiprogrammed system
c. to prevent overhead due to slow devices
d. to manage serial access to critical sections in processes

25. Interrupts allow more efficient use of the CPU because
a. the CPU can continue with code execution when slow peripherals are responding to an I/O request
b. polling would cause the CPU to spin-lock while waiting for a peripheral to respond
c. interrupt flags are checked inside the CPU without signicant impact on the fetch-decode-execute cycle
d. they occur very infrequently when compared with the frequency of the CPU executional cycle
e. none of the above

26. Scheduling the movements of the read/write head of a disk
a. is good because disks are expensive
b. is good because the disk is not rotating fast enough
c. is good because rotational latency increases the seek time
d. is good since rapid changes in direction of motion of the read write head means that it is stationary and not reading or writing
e. is only possible if we queue block requests

27. Possible causes for a process to relinquish access to the CPU are:
a. it has finished processing
b. it is preempted
c. it is blocked for an I/O operation
d. it has caused a security violation
e. none of the above

28. The wait-time for a process is
a. the amount of time a process needs to be in the running queue before it is completed
b. the time a process waits for all required resources to be allocated to it
c. the amount of time between arrival of a process in the ready queue and the moment that the process exits from the ready queue
d. the amount if time a process spends waiting in the ready queue
e. none of the above

29. A scheduling policy may attempt to optimize
a. the average throughput rate
b. the average turn-around time
c. CPU utilization
d. the average waiting time
e. none of the above

30. Non-preemptive scheduling algorithms
a. are a preferred class of algorithms to support real-time computing
b. rely on cooperative sharing mechanisms to multiplex the CPU
c. are most widely used because of their simplicity
d. rely on interrupts to multiplex the CPU
e. none of the above

31. FCFS CPU scheduling
a. maximizes throughput
b. maximizes equitability
c. minimizes wait time
d. is better for long processes
e. none of the above

32. A critical section
a. can be managed by enabling and disabling interrupts
b. is a trusted piece of code inside the OS kernel
c. is a section of code which can only be accessed by a single process at a time
d. once begun must be executed to completion
e. none of the above

33. A race condition
a. is a computing environment for benchmarking system programs
b. is a condition in which the order that processes execute certain sections of code determines the outcome of the computation
c. must execute as quickly as possible
d. none of the above

34. An acceptable solution to the critical section problem is achieved
a. if a critical section is free and a set of processes indicates a need to enter into the critical section, then only those processes competing for access to the critical section participate in the selection of a process to enter the critical section
b. if only one process at a time should be allowed into its critical section
c. if after a process requests entry into its critical section, only a bounded number of other processes is allowed to enter their related critical sections before the original process enters its critical section
d. if once a process attempts to enter its critical section, it cannot be postponed indefinitely even if no other process is in its critical section
e. by none of the above

35. A semaphore is
a. accessed only by two access routines
b. incremented by the signal (or V) operation
c. decremented by the wait (or P) operation
d. always a non-negative integer
e. none of the above

36. If the initial value of a semaphore is 1 and it is implemented as a non-negative integer then if the value of the semaphore is 0, we know
a. that another process is in its critical section and our process must wait
b. we can enter our critical section immediately
c. nothing, it depends on the code which we want to execute
d. we need to wait in a spin lock till the semaphore value returns to 1
e. none of the above

37. In the deadlock prevention strategy
a. we need to ensure that at least one of the necessary conditions for deadlock is false at all times
b. the Banker's algorihm is the best know implementation
c. mutual exclusion is not one of the necessary conditions that can be invalidated
d. prevention must be based on falsifying hold-and-wait, circular wait and no preemption of access to resources
e. none of the above

38. In a deadlock avoidance strategy
a. each process make its maximum claims known to memory manager when the process begins
b. the resource manager avoids deadlocks by staying in safe states
c. the Banker's algorithm is the best known implementation
e. none of the above

39. Which of the following are functions of the memory manager?
a. define a processes address space
b. allocate primary memory space to processes
c. minimize access times using a cost-effective amount of primary memory
d. map the process's address space into the allocated portion of primary memory
e. none of the above

Sample Questions for Part B

The textbook - Silberschatz, Galvin and Gagne, Operating System Concepts, Sixth Edition - has good examples of this type of written question. Some selected examples are:
Chapter 1: Exercises 1.2, 1.3
Chapter 2: Exercises 2.2, 2.3, 2.9
Chapter 3: Exercises 3.6, 3.7
Chapter 4: Exercises 4.2, 4.4
Chapter 6: Exercises 6.2, 6.3
Chapter 7: Exercises 7.1, 7.2, 7.3
Chapter 8: Exercises 8.8, 8.11, 8.13
Chapter 9: Exercises 9.7, 9.8, 9.10, 9.12
Chapter 10: Exercises 10.3, 10.5, 10.11, 10.14
Chapter 11: Exercises 11.1, 11.3, 11.10
Chapter 12: Exercises 12.1, 12.2, 12.4, 12.9
Chapter 13: Exercises 13.2, 13.4
Chapter 14: Exercises 14.1, 14.2, 14.9

Trevor Dix
Last modified: 23 May 2005

© Trevor Dix, Monash University, 2005

393 accesses since 10 Oct 2008, HTML cache rendered at 20070926:1120