1. In a broadcast and reduction on a balanced binary tree reduction is done in ______
Correct : A. recursive order
2. if "X" is the message to broadcast it initially resides at the source node
Correct : D. 0
3. logical operators used in algorithm are
Correct : C. both
4. Generalization of broadcast in Which each processor is
Correct : A. source as well as destination
5. The algorithm terminates in _____ steps
Correct : D. p-1
6. Each node first sends to one of its neighbours the data it need to....
Correct : A. broadcast
7. The second communication phase is a columnwise ______ broadcast of consolidated
Correct : A. all-to-all
8. All nodes collects _____ message corresponding to √p nodes to their respectively
Correct : A. √p
9. It is not possible to port ____ for higher dimensional network
Correct : A. algorithm
10. If we port algorithm to higher dimemsional network it would cause
Correct : B. contention
11. In the scatter operation ____ node send message to every other node
Correct : A. single
12. The gather Operation is exactly the inverse of _____
Correct : A. scatter operation
13. Similar communication pattern to all-to-all broadcast except in the_____
Correct : A. reverse order
14. Group communication operations are built using which primitives?
Correct : C. point to point
15. ___ can be performed in an identical fashion by inverting the process.
Correct : B. reduction
16. Broadcast and reduction operations on a mesh is performed
Correct : C. both a and b concurrently
17. Cost Analysis on a ring is
Correct : A. (ts + twm)(p - 1)
18. Cost Analysis on a mesh is
Correct : D. 2ts(sqrt(p) - 1) + twm(p - 1)
19. Communication between two directly link nodes
Correct : C. nearest neighbour communication
20. All-to-one communication (reduction) is the dual of ______ broadcast.
Correct : B. one-to-all
21. Which is known as Reduction?
Correct : A. all-to-one
22. Which is known as Broadcast?
Correct : B. one-to-all
23. The dual of all-to-all broadcast is
Correct : A. all-to-all reduction
24. All-to-all broadcast algorithm for the 2D mesh is based on the
Correct : B. ring algorithm
25. In the first phase of 2D Mesh All to All, the message size is ___
Correct : C. m
26. In the second phase of 2D Mesh All to All, the message size is ___
Correct : D. m*sqrt(p)
27. In All to All on Hypercube, The size of the message to be transmitted at the next step is ____ by concatenating the received message with their current data
Correct : A. doubled
28. The all-to-all broadcast on Hypercube needs ____ steps
Correct : C. log p
29. One-to-All Personalized Communication operation is commonly called ___
Correct : C. scatter operation
30. The dual of the scatter operation is the
Correct : C. both
31. In Scatter Operation on Hypercube, on each step, the size of the messages communicated is ____
Correct : B. halved
32. Which is also called "Total Exchange" ?
Correct : B. all-to-all personalized communication
33. All-to-all personalized communication can be used in ____
Correct : D. all of the above
34. In collective communication operations, collective means
Correct : A. involve group of processors
35. efficiency of data parallel algorithm depends on the
Correct : B. efficient implementation of the operation
36. All processes participate in a single ______ interaction operation.
Correct : A. global
37. subsets of processes in ______ interaction.
Correct : B. local
38. Goal of good algorithm is to implement commonly used _____ pattern.
Correct : A. communication
39. Reduction can be used to find the sum, product, maximum, minimum of _____ of numbers.
Correct : C. sets
40. source ____ is bottleneck.
Correct : A. process
41. only connections between single pairs of nodes are used at a time is
Correct : B. poor utilization
42. all processes that have the data can send it again is
Correct : A. recursive doubling
43. The ____ do not snoop the messages going through them.
Correct : A. nodes
44. accumulate results and send with the same pattern is...
Correct : D. reduction symmetric
45. every node on the linear array has the data and broadcast on the columns with the linear array algorithm in _____
Correct : A. parallel
46. using different links every time and forwarding in parallel again is
Correct : A. better for congestion
47. In a balanced binary tree processing nodes is equal to
Correct : A. leaves
48. In one -to- all broadcast there is
Correct : A. divide and conquer type algorithm
49. For sake of simplicity, the number of nodes is a power of
Correct : B. 2
50. Nides with zero in i least significant bits participate in _______
Correct : C. communication
51. every node has to know when to communicate that is
Correct : A. call the procedure
52. the procedure is disturbed and require only point-to-point _______
Correct : A. synchronization
53. Renaming relative to the source is _____ the source.
Correct : A. xor
54. Task dependency graph is ------------------
Correct : C. directed acyclic
55. In task dependency graph longest directed path between any pair of start and finish node is called as --------------
Correct : B. critical path
56. which of the following is not a granularity type
Correct : B. large grain
57. which of the following is a an example of data decomposition
Correct : A. matrix multiplication
58. which problems can be handled by recursive decomposition
Correct : C. divide and conquer problem
59. In this decomposition problem decomposition goes hand in hand with its execution
Correct : C. explorative decomposition
60. which of the following is not an example of explorative decomposition
Correct : D. quick sort
61. Topological sort can be applied to which of the following graphs?
Correct : D. directed acyclic graphs
62. In most of the cases, topological sort starts from a node which has __________
Correct : D. zero degree
63. Which of the following is not an application of topological sorting?
Correct : D. ordered statistics
64. In ------------task are defined before starting the execution of the algorithm
Correct : B. static task
65. which of the following is not the array distribution method of data partitioning
Correct : D. chunk
66. blocking optimization is used to improve temmporal locality for reduce
Correct : B. misses
67. CUDA thought that 'unifying theme' of every form of parallelism is
Correct : C. cuda thread
68. Topological sort of a Directed Acyclic graph is?
Correct : C. sometimes unique and sometimes not unique
69. threads being block altogether and being executed in the sets of 32 threads called a
Correct : A. thread block
70. True or False: The threads in a thread block are distributed across SM units so that each thread is executed by one SM unit.
Correct : A. true
71. When the topological sort of a graph is unique?
Correct : A. when there exists a hamiltonian path in the graph
72. What is a high performance multi-core processor that can be used to accelerate a wide variety of applications using parallel computing.
Correct : C. gpu
73. A good mapping does not depends on which following factor
Correct : D. task overhead
74. CUDA is a parallel computing platform and programming model
Correct : A. true
75. Which of the following is not a form of parallelism supported by CUDA
Correct : A. vector parallelism - floating point computations are executed in parallel on wide vector units
76. The style of parallelism supported on GPUs is best described as
Correct : B. simt - single instruction multiple thread
77. True or false: Functions annotated with the __global__ qualifier may be executed on the host or the device
Correct : A. true
78. Which of the following correctly describes a GPU kernel
Correct : B. all thread blocks involved in the same computation use the same kernel
79. a code known as grid which runs on GPU consisting of a set of
Correct : D. thread block
80. which of the following is not an parallel algorithm model
Correct : C. task model
81. Having load before the store in a running program order, then interchanging this order, results in a
Correct : C. war hazards
82. model based on the passing of stream of data through process arranged in a succession is called as
Correct : A. producer consumer model
83. When instruction i and instruction j are tends to write the same register or the memory location, it is called
Correct : B. output dependence
84. Multithreading allowing multiple-threads for sharing the functional units of a
Correct : B. single processor
85. Allowing multiple instructions for issuing in a clock cycle, is the goal of
Correct : C. multiple-issue processors
86. OpenGL stands for:
Correct : B. open graphics library
87. which of the following is not an advantage of OpenGL
Correct : D. it is not a cross-platform api,
88. work pool model uses ---------------- approach for task assignment
Correct : B. dynamic
89. which of the following is false regarding data parallel model
Correct : D. dynamic mapping is done
90. which of the following are methods for containing interaction overheads
Correct : D. all the above
91. which of the following are classes of dynamic mapping centralized method
Correct : C. both a and b
92. which of the following is not scheme for static mapping
Correct : D. self scheduling
93. A pipeline is like ....................
Correct : A. an automobile assembly line
94. Data hazards occur when .....................
Correct : B. pipeline changes the order of read/write access to operands
95. Systems that do not have parallel processing capabilities are
Correct : A. sisd
96. How does the number of transistors per chip increase according to Moore ´s law?
Correct : D. exponentially
97. Parallel processing may occur
Correct : C. both[a] and [b]
98. Execution of several activities at the same time.
Correct : B. parallel processing
99. Cache memory works on the principle of
Correct : C. locality of reference ??
100. SIMD represents an organization that ______________.
Correct : C. ?? ?? ?? includes many processing units under the supervision ?? ?? ?? ?? ?? of a common control unit