You are on page 1of 1

13392 13392 13392 13392 13392 13392 13392 13392 13392

13392 Roll No.


13392 13392 13392 13392 13392 Total13392
No. of Pages
13392 : 01 13392

Total No. of Questions : 08


M.Tech.(IT) / (CSE Engg.) / (E-Security) (Sem.–1)
13392 13392 13392 13392 13392 13392 13392 13392 13392
ADVANCED COMPUTER ARCHITECTURE
Subject Code : CS-505
Paper ID : [E0685]
13392 13392 13392 13392 13392 13392 13392 13392 13392

Time : 3 Hrs. Max. Marks : 100

INSTRUCTION TO CANDIDATES :

m
13392 13392 13392 13392 13392 13392 13392 13392 13392
1. Attempt any FIVE questions out of EIGHT questions.

o
2. Each question carries T WENT Y marks.

1. a) Explain

.r c
13392 13392 13392 the various methods by which
13392 13392 data level parallelism is13392
13392 obtained? 13392 13392

b) What is pipelining? Name different stages of pipelined architecture. Briefly

13392 2.
13392 What do
13392you understand
13392

pe
describe basic performance issues in pipelining.
by static and dynamic
13392

o
scheduling?13392
13392 m
Explain with13392
suitable 13392

a .r c
examples the Tomasulo’s algorithm for MIPS processor.
3. a) Define the terms ILP. Discuss how compiler techniques can be exploited for
13392 13392 achieving
13392

r p
ILP? 13392 13392 13392

e
13392 13392 13392

b
b) Discuss their advantages and challenges of parallel processing.

13392 13392
4. 13392
a) What 13392
is a static branch 13392
prediction?
a p
c) What are forbidden and permissible latencies? Give example.
13392
What are the 13392
various methods available13392
for static 13392

rp
branch prediction?
b) Define the terms : Name dependences, control dependence and data dependence,
13392 13392

5.
loop dependence. 13392
13392 13392

b 13392 13392 13392

Why, there is a need to detect loop dependences? How does the compiler detect it?
a) Explain in detail about hardware multithreading techniques.
13392

13392 13392 13392 13392 13392 13392 13392 13392 13392


b) What are VLIW processors? Discuss the various problems associated with the
VLIW processor and measures for their mitigation.
6. Explain in detail the centralized shared memory architectures and symmetric shared
13392 13392 13392 13392 13392 13392 13392 13392 13392
memory multiprocessors.
7. a) How performance of parallel processors is measured? With relevant graphs,
discuss the performance of distributed shared memory multiprocessors.
13392 13392 13392 13392 13392 13392 13392 13392 13392
b) What is cache coherence problem and when do you say a memory system is
coherent? What are cache coherence protocols?
8. Write notes on the following :
13392 13392 13392 13392 13392 13392 13392 13392 13392
a) Superscalar architectures
b) Intel Multi-core Architecture

13392 1 | M-35406
13392 13392 13392 13392 13392 13392 (S9)-1696
13392 13392

13392 13392 13392 13392 13392 13392 13392 13392 13392

You might also like