Question 1

_________ is Callable from the device only

Accepted Answer

A)  _host_
B)  __global__
C)  _device_
D)  none of above C

Question 2

______ is Callable from the host

Accepted Answer

A)  _host_
B)  __global__
C)  _device_
D)  none of above A

Question 3

Host codes in a CUDA application can Initialize a device

Accepted Answer

A) True 
 B)False  True

Question 4

The time lost due to branch instruction is often referred to as _____.

Accepted Answer

A)  latency
B)  delay
C)  branch penalty
D)  none of the above

Question 5

______ is Callable from the host

Accepted Answer

A)  _host_
B)  __global__
C)  _device_
D)  none of above

Question 6

The computer cluster architecture emerged as an alternative for ____.

Accepted Answer

A)  isa
B)  workstation
C)  super computers
D)  distributed systems

Question 7

Host codes in a CUDA application can Transfer data to and from the device

Accepted Answer

A) True 
 B)False

Question 8

Any condition that causes a processor to stall is called as _____.

Accepted Answer

A)  hazard
B)  page fault
C)  system error
D)  none of the above

Question 9

A grid is comprised of ________ of threads.

Accepted Answer

A)  block
B)  bunch
C)  host
D)  none of above

Question 10

Calling a kernel is typically referred to as _________.

Accepted Answer

A)  kernel thread
B)  kernel initialization
C)  kernel termination
D)  kernel invocation

Question 11

In CUDA, a single invoked kernel is referred to as a _____.

Accepted Answer

A)  block
B)  tread
C)  grid
D)  none of above

Question 12

a solution of the problem in representing the parallelismin algorithm is

Accepted Answer

A)  cud
B)  pta
C)  cda
D)  cuda

Question 13

The kernel code is dentified by the ________qualifier with void return type

Accepted Answer

A)  _host_
B)  __global__
C)  _device_
D)  void

Question 14

A block is comprised of multiple _______.

Accepted Answer

A)  treads
B)  bunch
C)  host
D)  none of above

Question 15

Host codes in a CUDA application can not Reset a device

Accepted Answer

A) True 
 B)False

Question 16

the BlockPerGrid and ThreadPerBlock parameters are related to the ________ model supported by CUDA.

Accepted Answer

A)  host
B)  kernel
C)  threadabstraction
D)  none of above

Question 17

CUDA offers the Chevron Syntax to configure and execute a kernel.

Accepted Answer

A) True 
 B)False

Question 18

CUDA supports ____________ in which code in a single thread is executed by all other threads.

Accepted Answer

A)  tread division
B)  tread termination
C)  thread abstraction
D)  none of above

Question 19

The kernel code is only callable by the host

Accepted Answer

A) True 
 B)False

Question 20

Host codes in a CUDA application can not Deallocate memory on the GPU

Accepted Answer

A) True 
 B)False

Exam 1: Understanding Cuda Kernel Code and Host GPU Interactions

_________ is Callable from the device only

______ is Callable from the host

Host codes in a CUDA application can Initialize a device

The time lost due to branch instruction is often referred to as _____.

______ is Callable from the host

The computer cluster architecture emerged as an alternative for ____.

Host codes in a CUDA application can Transfer data to and from the device

Any condition that causes a processor to stall is called as _____.

A grid is comprised of ________ of threads.

Calling a kernel is typically referred to as _________.

In CUDA, a single invoked kernel is referred to as a _____.

a solution of the problem in representing the parallelismin algorithm is

The kernel code is dentified by the ________qualifier with void return type

A block is comprised of multiple _______.

Host codes in a CUDA application can not Reset a device

the BlockPerGrid and ThreadPerBlock parameters are related to the ________ model supported by CUDA.

CUDA offers the Chevron Syntax to configure and execute a kernel.

CUDA supports ____________ in which code in a single thread is executed by all other threads.

The kernel code is only callable by the host

Host codes in a CUDA application can not Deallocate memory on the GPU