High Performance Computing Set 12 MCQs

Q1 | the BlockPerGrid and ThreadPerBlock parameters are related to the ________ model supported by CUDA.

host

kernel

thread abstraction

none of above

Q2 | _________ is Callable from the device only

_host_

__global__

_device_

none of above

Q3 | ______ is Callable from the host

_host_

__global__

_device_

none of above

Q4 | ______ is Callable from the host

_host_

__global__

_device_

none of above

Q5 | Cache memory works on the principle of

locality of data

locality of memory

locality of reference

locality of reference & memory

Q6 | SIMD represents an organization that ______________.

refers to a computer system capable of processing several programs at the same time.

represents organization of single computer containing a control unit, processor unit and a memory unit.

includes many processing units under the supervision of a common control unit

none of the above.

Q7 | Select different aspects of parallelism

server applications utilize high aggregate network bandwidth

scientific applications typically utilize high processing and memory system performance

all of the above

data intensive applications utilize high aggregate throughput

Q8 | Select correct answer: DRAM access times have only improved at the rate of roughly % per year over this interval.

20

40

50

10

Q9 | Analyze, if the second instruction has data dependencies with the first, but the third instruction does not, the first

out-of-order

both of the above

none of the above

in-order

Q10 | Select the parameters which captures Memory system performance

bandwidth

both of the above

none of the above

latency

Q11 | Consider the example of a fire- hose. If the water comes out of the hose five seconds after the hydrant is turned on. Once the water starts flowing, if the hydrant delivers water at the rate of 15 gallons/second. Analyze the bandwidth and latency.

bandwidth: 5*15 gallons/second and latency: 15 seconds

bandwidth: 15 gallons/second and latency: 5 seconds

bandwidth: 3 gallons/second and latency: 5 seconds

bandwidth: 5 gallons/second and latency: 15 seconds

Q12 | Select alternate approaches for Hiding Memory Latency

multithreading

spatial locality

all of the above

prefeching

Q13 | Select which clause in OpenMP is similar to the private, except values of variables are initialized to corresponding values before the

firstprivate

shared

all of the above

private

Q14 | Which of the following projects of Blue Gene is not in development?

blue gene / m

blue gene / p

blue gene / q

blue gene / l

Q15 | A decomposition can be illustrated in the form of a directed graph with nodes corresponding to tasks and edges indicating that the result of one task is required for processing the next. Such graph is called as

task dependency graph

task interaction graph

process interaction graph

process dependency graph

Q16 | In which case, the owner computes rule implies that the output is computed by the process to which the output data is assigned?

output data decomposition

both of the above

none of the above

input data decomposition

Q17 | Select relevant task characteristics from the options given below:

task sizes

size of data associated with tasks

all of the above

task generation

Q18 | A classic example of game playing - each 15 puzzle board is the example of

dynamic task generation

none of the above

all of the above

static task generation

Q19 | Which model is equally suitable to shared-address- space or message- passing paradigms, since the interaction is naturally two ways.

master slave model

data parallel model

producer consumer or pipeline model

work pool model

Q20 | In which type of the model, tasks are dynamically assigned to the processes for balancing the load?

master slave model

data parallel model

producer consumer or pipeline model

work pool model

Q21 | Select the appropriate stage of GPU Pipeline which receives commands from CPU and also pulls geometry information from system memory.

vertex processing

memory interface

host interface

pixel processing

Q22 | In all-to-one reduction, data items must be combined piece-wise and the result made available at a processor.

last

target

n-1

first

Q23 | Analyze the Cost of Scatter and Gather .

t=ts log p + tw m (p-1)

t=ts log p - tw m (p-1)

t=tw log p - ts m (p-1)

t=tw log p + ts m (p-1)

Q24 | All-to-all personalized communication is also known as .

total exchange

both of the above

none of the above

partial exchange

Q25 | All-to-all personalized communication is performed independently in each row with clustered messages of size on a mesh.

p

m√p

p√m

m