Computer Science Engineering (CSE) Exam  >  Computer Science Engineering (CSE) Tests  >  Operating System  >  Test: CPU & I/O Scheduling- 2 - Computer Science Engineering (CSE) MCQ

Test: CPU & I/O Scheduling- 2 - Computer Science Engineering (CSE) MCQ


Test Description

20 Questions MCQ Test Operating System - Test: CPU & I/O Scheduling- 2

Test: CPU & I/O Scheduling- 2 for Computer Science Engineering (CSE) 2024 is part of Operating System preparation. The Test: CPU & I/O Scheduling- 2 questions and answers have been prepared according to the Computer Science Engineering (CSE) exam syllabus.The Test: CPU & I/O Scheduling- 2 MCQs are made for Computer Science Engineering (CSE) 2024 Exam. Find important definitions, questions, notes, meanings, examples, exercises, MCQs and online tests for Test: CPU & I/O Scheduling- 2 below.
Solutions of Test: CPU & I/O Scheduling- 2 questions in English are available as part of our Operating System for Computer Science Engineering (CSE) & Test: CPU & I/O Scheduling- 2 solutions in Hindi for Operating System course. Download more important topics, notes, lectures and mock test series for Computer Science Engineering (CSE) Exam by signing up for free. Attempt Test: CPU & I/O Scheduling- 2 | 20 questions in 55 minutes | Mock test for Computer Science Engineering (CSE) preparation | Free important questions MCQ to study Operating System for Computer Science Engineering (CSE) Exam | Download free PDF with solutions
Test: CPU & I/O Scheduling- 2 - Question 1

For the processes listed in the following table, which of the following scheduling schemes will give the lowest average turnaround time?

Process    Arrival Time    Processing Time
  A                  0                      3
  B                 1                       6
  C                 4                       4
  D                 6                       2

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 1

Turnaround time is the total time taken between the submission of a program/process/thread/task (Linux) for execution and the return of the complete output to the customer/user. Turnaround Time = Completion Time - Arrival Time. FCFS = First Come First Serve (A, B, C, D) SJF = Non-preemptive Shortest Job First (A, B, D, C) SRT = Shortest Remaining Time (A(3), B(1), C(4), D(2), B(5)) RR = Round Robin with Quantum value 2 (A(2), B(2), A(1),C(2),B(2),D(2),C(2),B(2)  

Test: CPU & I/O Scheduling- 2 - Question 2

Which of the following is FALSE about SJF (Shortest Job First Scheduling)?
S1: It causes minimum average waiting time
S2: It can cause starvation

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 2
  1. Both SJF and Shortest Remaining time first algorithms may cause starvation. Consider a situation when long process is there in ready queue and shorter processes keep coming.
  2. SJF is optimal in terms of average waiting time for a given set of processes, but problems with SJF is how to know/predict time of next job.
1 Crore+ students have signed up on EduRev. Have you? Download the App
Test: CPU & I/O Scheduling- 2 - Question 3

Consider an arbitrary set of CPU-bound processes with unequal CPU burst lengths submitted at the same time to a computer system. Which one of the following process scheduling algorithms would minimize the average waiting time in the ready queue?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 3

Turnaround time is the total time taken by the process between starting and the completion and waiting time is the time for which process is ready to run but not executed by CPU scheduler. As we know, in all CPU Scheduling algorithms, shortest job first is optimal i.ie. it gives minimum turn round time, minimum average waiting time and high throughput and the most important thing is that shortest remaining time first is the pre-emptive version of shortest job first. shortest remaining time first scheduling algorithm may lead to starvation because If the short processes are added to the cpu scheduler continuously then the currently running process will never be able to execute as they will get pre-empted but here all the processes are arrived at same time so there will be no issue such as starvation. So, the answer is Shortest remaining time first, which is answer (A).

Test: CPU & I/O Scheduling- 2 - Question 4

Consider the following processes, with the arrival time and the length of the CPU burst given in milliseconds. The scheduling algorithm used is preemptive shortest remaining-time first.

z5

The average turn around time of these processes is ___________ milliseconds.   Note : This question was asked as Numerical Answer Type.

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 4

PreEmptive Shortest Remaining time first scheduling, i.e. that processes will be scheduled on the CPU which will be having least remaining burst time( required time at the CPU). The processes are scheduled and executed as given in the below Gantt chart.

Turn Around Time(TAT) = Completion Time(CT) - Arrival Time(AT) TAT for P1 = 20 - 0 = 20 TAT for P2 = 10 - 3 = 7 TAT for P3 = 8- 7 = 1 TAT for P4 = 13 - 8 = 5 Hence, Average TAT = Total TAT of all the processes / no of processes = ( 20 + 7 + 1 + 5 ) / 4 = 33 / 4 = 8.25   Thus, A is the correct choice.

Test: CPU & I/O Scheduling- 2 - Question 5

Assume every process requires 3 seconds of service time in a system with single processor. If new processes are arriving at the rate of 10 processes per minute, then estimate the fraction of time CPU is busy in system?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 5

10 processes -> 1 min
1 process-> 1/10 min = 6 sec (Arrival rate)
Each process -> 3 sec service time 3/6 * 100 = 50% of time CPU is busy.

Test: CPU & I/O Scheduling- 2 - Question 6

When an interrupt occurs, an operating system

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 6

Scheduler decides that the interrupted process will complete execution or some other process will be executed. If the interrupt signaled an I/O completion event, and at the same time a high priority process came into Ready state then the scheduler block the interrupted process and dispatch the high priority process in the running state. If low priority process comes into Ready state then scheduler dispatch the interrupted process. Hence, D is correct.

Test: CPU & I/O Scheduling- 2 - Question 7

Consider the following CPU processes with arrival times (in milliseconds) and length of CPU bursts (in milliseconds) except for process P4 as given below:
Process    Arrival Time    Burst Time
P1                  0                  5
P2                  1                  1
P3                  3                  3
P4                  4                  x
If the average waiting time across all processes is 2 milliseconds and pre-emptive shortest remaining time first scheduling algorithm is used to schedule the processes, then find the value of x ?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 7

If we take value of x is 2, then we have gantt chart as

So, completion time of P1, P2, P3, and P4 are 6, 2, 11, and 8 respectively. Turn around time of P1, P2, P3, and P4 are 6, 1, 8, and 4 respectively. Waiting time of P1, P2, P3, and P4 are 1, 0, 5, and 2 resectively. Therefore, average waiting time = (1+0+5+2) / 4 = 8/2 = 2 Option (B) is correct.

Test: CPU & I/O Scheduling- 2 - Question 8

Consider the following four processes with the arrival time and length of CPU burst given in milliseconds :

49

The average waiting time for preemptive SJF scheduling algorithm is __________.

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 8

First we will make gantt chart of given process then we will calculate turn around time and waiting time of individual process.

Now we have to calculate average waiting time for schedule:avg waiting time = wt(P1 + P2 + P3 + P4 )/number of process.    ie.            

(9 + 0 + 15 + 2) / 4
= 26 / 4
= 6.5

Test: CPU & I/O Scheduling- 2 - Question 9

Which module gives control of the CPU to the process selected by the short - term schedular ?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 9

There are three type of scheduler:

  1. Short-term scheduler
  2. Mid-term scheduler
  3. Long-term scheduler

Dispatcher is responsible for handovering the control of CPU to the newly selected process by the Short-term scheduler. Refer: Process Scheduler So, option (A) is correct.

Test: CPU & I/O Scheduling- 2 - Question 10

Which of the following scheduling algorithms may cause starvation ? a. First-come-first-served b. Round Robin c. Priority d. Shortest process next e. Shortest remaining time first

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 10
  1. In First Come First Serve(FCFS) if a process with a very large Burst Time comes before other processes, the other process will have to wait for a long time but it is clear that other process will definitely get their chance to execute, so it will not suffer from starvation.
  2. In Round Robin there is a fixed time quant and every process will get their chance to be executed, so no starvation is here.
  3. In Priority based scheduling if higher priority process keep on coming then low priority process will suffer from starvation.
  4. In Shortest Job First(SJF) if process with short process time keep on coming continuously then process with higher burst time will do wait and suffer from starvation.
  5. In Shortest remaining time first(SRTF) process with shortest burst time will execute first because of this process with high burst time may suffer from starvation.

So, option (C) is correct.

Test: CPU & I/O Scheduling- 2 - Question 11

Which of the following statement is true?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 11

Jitter is the variation / displacement between the signals or data being sent. Hard real operating systems deal with more sensitive systems which require strict time deadlines like engine control systems, satellite launching systems etc while soft real operating systems do not require strict timing constraints and a bit delay is permissible, like mobile phones, online database systems. So hard real operating systems require minimised jitter.

Test: CPU & I/O Scheduling- 2 - Question 12

Consider three CPU intensive processes P1, P2, P3 which require 20, 10 and 30 units of time, arrive at times 1, 3 and 7 respectively. Suppose operating system is implementing Shortest Remaining Time first (preemptive scheduling) algorithm, then _____ context switches are required (suppose context switch at the beginning of Ready queue and at the end of Ready queue are not counted).

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 12

40

There are 3 context switch. So. option (A) is correct.

Test: CPU & I/O Scheduling- 2 - Question 13

Five jobs A, B, C, D and E are waiting in Ready Queue. Their expected runtimes are 9, 6, 3, 5 and x respectively. All jobs entered in Ready queue at time zero. They must run in _____ order to minimize average response time if 3 < x < 5.

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 13

We will solve it by minimizing the avg Waiting Time and Take x = 4 BT = Burst or Execution time CT = completion time P = Process AWT = Average Waiting Time AT = Arrival time WT = Waiting time

We get minimum average waiting time from the sequence given in option (B)

Test: CPU & I/O Scheduling- 2 - Question 14

Which of the following is not an optimization criterion in the design of a CPU scheduling algorithm?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 14

Minimum CPU utilization is not an optimization criterion as various optimization techniques and scheduling algorithms are used to bring the best CPU performance.

Test: CPU & I/O Scheduling- 2 - Question 15

Below is the precedence graph for a set of tasks to be executed on a parallel processing system S.

What is the efficiency of this precedence graph on S if each of the tasks T1, T2, T3,....T8 takes the same time and the system S has five processors?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 15

It can be seen that along with the sequential execution of T1 and T2, (T3, T6), (T4, T7) and (T5, T8), all these three processes can be executed in parallel. So total number of processes that can be executed in 4 units time using 5 available processors = 5*4 = 20 But here processes that are executing in 4 units time = 8 Throughput = 8/20 * 100 = 40% So, option (B) is correct.

Test: CPU & I/O Scheduling- 2 - Question 16

A scheduling Algorithm assigns priority proportional to the waiting time of a process. Every process starts with priority zero (lowest priority). The scheduler reevaluates the process priority for every ‘T’ time units and decides next process to be scheduled. If the process have no I/O operations and all arrive at time zero, then the scheduler implements _________ criteria.

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 16

A scheduling Algorithm assigns priority proportional to the waiting time of a process. Every process starts with priority zero (lowest priority). The scheduler reevaluates the process priority for every ‘T’ time units and decides next process to be scheduled. If the process have no I/O operations and all arrive at time zero, then the scheduler implements Round Robin Scheduling criteria. So, option (B) is correct.

Test: CPU & I/O Scheduling- 2 - Question 17

In a system using single processor, a new process arrives at the rate of six processes per minute and each such process requires seven seconds of service time. What is the CPU utilization?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 17

Number of processes per minute = 6 Burst time of each process = 7 secs
CPU utilization time within a minute = 6*7 = 42 secs
% CPU utilization = useful time / total time * 100
= (42/60) * 100 
 70%
Option (A) is correct.

Test: CPU & I/O Scheduling- 2 - Question 18

Consider the following justifications for commonly using the two-level CPU scheduling: I. It is used when memory is too small to hold all the ready processes. II. Because its performance is same as that of the FIFO. III. Because it facilitates putting some set of processes into memory and a choice is made from that. IV. Because it does not allow to adjust the set of in-core processes. Which of the following is true ?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 18

The two-level CPU scheduling is used when memory is too small to hold all the ready processes because it facilitates putting some set of processes into memory and a choice is made from that. So, option (D) is correct.

Test: CPU & I/O Scheduling- 2 - Question 19

Which of the following statements is not true for Multi Level Feedback Queue processor scheduling algorithm?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 19

For Multi Level Feedback Queue processor scheduling algorithm:

  • Queues have different priorities
  • Each queue may have different scheduling algorithm
  • Processes are not permanently assigned to a queue.
  • This algorithm can be configured to match a specific system under design
Test: CPU & I/O Scheduling- 2 - Question 20

In which of the following scheduling criteria, context switching will never take place ?

Detailed Solution for Test: CPU & I/O Scheduling- 2 - Question 20

In Non - preemptive algorithms context switching will never take place because it doesn't allow to switch the process until it is completed fully. So, option (C) is correct.

10 videos|99 docs|33 tests
Information about Test: CPU & I/O Scheduling- 2 Page
In this test you can find the Exam questions for Test: CPU & I/O Scheduling- 2 solved & explained in the simplest way possible. Besides giving Questions and answers for Test: CPU & I/O Scheduling- 2, EduRev gives you an ample number of Online tests for practice

Top Courses for Computer Science Engineering (CSE)

10 videos|99 docs|33 tests
Download as PDF

Top Courses for Computer Science Engineering (CSE)