HPC May
HPC May
HPC May
8
23
P-558 [Total No. of Pages : 2
ic-
[6004]-493
tat
7s
B.E. (Computer Engineering)
9:2
02 91
High Performance Computing
4:2
0
(2019 Pattern) (Semester - VIII) (410250)
31
7/0 13
0
Time : 2½ Hours] [Max. Marks : 70
6/2
.23 GP
8
2) Neat diagrams must be drawn wherever necessary.
C
23
3) Figures to the right indicate full marks.
ic-
16
tat
8.2
7s
Q1) a) Explain with diagram One-to-all broadcast on an eight-node ring with
.24
9:2
91
recursive doubling technique. Node 0 is the source of the broadcast.
49
4:2
Also Explain all to one reduction with node 0 as destination. [7]
30
31
MPI. [6]
6/2
GP
OR
80
8
23
.23
tat
first two steps and last communication step). [7]
8.2
7s
.24
9:2
4:2
example? [7]
7/0
OR
.24
49
P.T.O.
Q4) a) What is granularity? What are effects of granularity on performance of
8
parallel systems? [7]
23
ic-
b) Explain various sources of overhead in parallel systems? [6]
tat
c) Explain “Scaling Down (downsizing)” a parallel system with example.
7s
[4]
9:2
02 91
4:2
0
Q5) a) What is CUDA? Explain different programming languages support in
31
7/0 13
CUDA. Discuss any three applications of CUDA. [8]
0
b) Describe processing flow of CUDA-C program with diagram. [6]
6/2
.23 GP
Kernel. [4]
80
8
C
23
OR
ic-
16
tat
8.2
7s
b) What is block dimension and grid dimension in CUDA? Write a CUDA
.24
9:2
kernel for addition of two vectors and explain how it will calculate addition
91
using threads. [6]
49
4:2
30
8
23
transportation. [8]
.23
tat
8.2
7s
9:2
OR
91
49
4:2
[4]
16
8.2
.24
49
[6004]-493 2