Semester : SEMESTER 7
Subject : Computer System Architecture
Year : 2018
Term : DECEMBER
Branch : COMPUTER SCIENCE AND ENGINEERING
Scheme : 2015 Full Time
Course Code : CS 405
Page:3
16
17
18
19
a)
b)
(a)
(b)
a)
b)
a)
b)
R7954 Pages: 3
v) Let the pipeline clock period be t=20ns. Determine the throughput of
the pipeline.
Explain full-map directory based protocol.
What do you mean by dimension order routing? Consider a 16 node hypercube
network. Based on E-cube routing algorithm, show how to route a message
from 0010 to 1001. Find all intermediate nodes on routing path.
PART D ۱
Answer any two full questions, each carries 12 marks.
Explain the Tomasulo’s algorithm for the dynamic instruction scheduling.
Explain the concept of in-order issue and out-of-order issue with respect to
superscalar processor.
Explain any three latency hiding techniques used in distributed shared memory
multi computers.
Write a short note on fine-grain parallelism.
Explain static branch prediction strategy and dynamic branch prediction
strategy.
With a neat diagram explain the architecture of ETL/EM-4 dataflow
architecture.
೫% ೫%
7೩63083
(6)
(4)
(5)
(5)
(7)
(9)
(3)
(6)
(6)