Parallel and Distributed Computing Unit - 3 mcq

 

 

1. A sequential algorithm is evaluated by its
(A) Compile time
(B) Runtime
(C) Both A & B
(D) None of these

Correct option is B

2. The parallel runtime of a program depends on
(A) Input Size
(B) Number of processors
(C) Communication parameters of the machine
(D) All of these

Correct option is D

3. The time from the start of the first processor to the stopping time of the last processor in a parallel ensemble is called
(A) Wall clock time
(B) Execution time
(C) Processing time
(D) None of these

Correct option is A

4. If I use two processors, shouldn’t my program run twice as fast?
(A) Yes
(B) No

Correct option is B

5. Serial runtime of a program is
(A) Time that elapses from the moment the first processor starts to the moment the last processor finishes execution.
(B) Time elapsed between the beginning and the end of its execution on a sequential computer.
(C) Both A & B
(D) None of the above

Correct option is B

6. Total overhead is
(A) The total time spend by all processors combined in non-useful work.
(B) The total time spend by all processors combined in useful work.
(C) The total time collectively spent by all the processing elements.
(D) Both B & C

Correct option is A

7. The overhead function (To) is therefore given by
(A) T_o = pT_p - T_s
(B) T_o = pT_p + T_s
(C) T_o = pT_s - T_p
(D) T_o = pT_s + T_p

Correct option is A

8. The ratio of the time taken to solve a problem on a single processor to the time required to solve the same problem on a parallel computer with p identical processing elements

(A) SpeedHigh
(B) SpeedUp
(C) SpeedFast
(D) None of these

Correct option is B

9. Consider the problem of adding n numbers, If n is a power of two, we can perform this operation in ___ steps by propagating partial sums up a logical binary tree of processors.
(A) n² log n
(B) n log n
(C) log n
(D) None of these

Correct option is C

10. Speedup can be as
(A) High as 1
(B) Low as 0
(C) Can be more than 1
(D) Can be less than 0

Correct option is B

11. A speedup greater than p is possible only if each processing element spends less than
(A) Ts/p
(B) p/Ts
(C) Ts × p
(D) None of these

Correct option is A

12. Cost of a Parallel System given by
(A) p x Tp
(B) p / Tp
(C) Tp / p
(D) p x p x Tp

Correct option is A

13. Using fewer than the maximum possible number of processing elements to execute a parallel algorithm.
(A) Scaling down a parallel system
(B) Scaling up a parallel system
(C) Scaling down a serial system
(D) Scaling up a serial system

Correct option is A

14. The number of processing elements decreases by a factor of n/p, then the computation at each processing element increases by a factor of __
(A) n/p
(B) n*p
(C) p/n
(D) None of these

Correct option is A

15. The efficiency of a parallel program can be written as
(A) S/p
(B) Ts / pT_P
(C) Both A & B
(D) None of these

Correct option is C

16. The maximum number of tasks that can be executed simultaneously at any time in a parallel algorithm is called its
(A) Degree of concurrency
(B) Isoefficiency Function
(C) None of these

Correct option is A

17. If the metric is speed, the order of algorithm is
(A) A1 > A3 > A4 > A2
(B) A3 > A1 > A4 > A2
(C) A1 > A3 > A4 > A2
(D) A3> A4 > A1 > A2

Correct option is D

18. In terms of efficiency, the order of algorithm
(A) A2 > A4 > A3 > A1
(B) A2 > A3 > A4 > A1
(C) A3 > A2 > A1 > A4
(D) A3 > A2 > A4 > A1

Correct option is A

19. If the serial runtime of a computation can be divided
(A) Totally parallel component
(B) Totally parallel component
(C) Both A & B
(D) None of these

Correct option is C

20. The serial fraction f of a parallel program is defined as
(A) f = T_ser / W
(B) f = T_ser X W
(C) f = W / T_ser
(D) None of these

Correct option is A