Vous êtes sur la page 1sur 30

Use Of GPU

Computing For
Big Data
Analytics
SRIYAM SINGH
M150501EC

CONTENTS

INTRODUCTION
LITERATURE REVIEW
PROJECT PHASES
REFERENCES

INTRODUCTION
GPU
architectures
are
becoming
increasingly
programmable, offering the potential for dramatic
speedups for a variety of general-purpose applications
compared to contemporary general-purpose processors
(CPUs). This project uses NVIDIAs C-like CUDA language
and an engineering sample of their recently introduced
Quadro K6000 GPU to explore the effectiveness of GPUs
for Big Data Analytics, and describes some specific coding
algorithms that further improve their performance on the
CPU-GPU combination.

LITERATURE REVIEW

why is there
such buzz around

BIG DATA?

... How big is BIG?

What is it that we are


really talking about?

Big Data is the frontier of a firm's


ability to store, process, and access
(SPA) all the data it needs to operate
effectively, make decisions, reduce
risks, and serve customers.
-- Forrester

Bo

rin

Big Data is the frontier of a firm's


ability to store, process, and access
(SPA) all the data it needs to operate
effectively, make decisions, reduce
risks, and serve customers.

g!

-- Forrester

Big data is data that exceeds the


processing capacity of conventional
database systems. The data is too big,
moves too fast, or doesn't fit the
strictures of your database
architectures. To gain value from this
data, you must choose an alternative
-- OReilly
way to process it.

Big data is data that exceeds the


processing capacity of conventional
database systems. The data is too big,
moves too fast, or doesn't fit the
strictures of your database
architectures. To gain value from this
data, you must choose an alternative
-- OReilly
way to process it.

y
n
a
m
o
o
T
s
d
r
o
w

Big data is the data characterized by 3


attributes: volume, variety and
velocity.
-- IBM

INTEREST
IN
G

Big data is the data characterized by 3


attributes: volume, variety and
velocity.
-- IBM

Lets look at

Big Data

in a different way.

Big Data is not about the size of the data,


its about the value within the data.

So, what is

value?

Market Intelligence?
Business Intelligence?
Public Intelligence?

How do we
mine (and mind)
all this data?

What is GPU?
It is a processor optimized for 2D/3D graphics, video,
and display.
It is highly parallel, highly multithreaded
multiprocessor optimized for visual computing.
It serves as both a programmable graphics
processor and a scalable parallel computing
platform.
It works along with CPU.

GPU vs CPU
A GPU is tailored for highly parallel operation
while a CPU executes programs serially.
For this reason, GPUs have many parallel
execution units while CPUs have few
execution units and higher clock speeds.
GPUs have significantly faster and more
advanced memory interfaces as they need to
shift around a lot more data than CPUs.

GPU vs CPU

Programming Flow

Anatomy of a CUDA C/C++ Application

Kernel Execution

PROJECT PHASES
PHASE 1

Comparing CPU and GPU with a simple code.


PHASE 2

Studying CUDA C , CuPy and BIG DATA analysis techniques.


Implimenting a solution of a problem in any one domain of Big Data and comparing
the performance of CPU and GPU.
PHASE 3

Implimenting the same problem's solution with different algorithms.

PHASE 4

Suggesting an optimal algorithm .


Suggesting possible improvements in existing algorithms.
Disscussing advantages and inefficiencies of CUDA programming
model.

PHASE 1 (RESULT)
MATRI GPU
X
TIME(
SIZE
ms)

CPU
TIME(
ms)

3x3

.23

6x6

.25

10

10x10

.58

31

100x1
00

1.99

839

150x1
50

5.87

1408

REFERENCES
[1] What is GPU Computing?
(http://www.nvidia.com/object/GPU_Computing.html)
[2] http://supercomputingblog.com/cuda/practical-applications-for-cuda
[3] Jeff Kelly. Big data vendor revenue and market forecast. Wikibon, 2014.
[4]Jacobs, A. The pathologies of big data. ACM Commun. 2009
[5]Wu,X.Zhu, X.; Wu, G.-Q.; Ding, W. Data mining with big data. IEEE Trans.
Data Eng.2014.
[5] Zhigao Zheng , Ping Wang. Real-Time Big Data Processing Framework:
Challenges
and Solutions. 1 Nov,2015

THANK YOU!!!

Vous aimerez peut-être aussi