Transcript
1
A DATA MAPPING STRATEGY FOR PARALLEL DATA MINING NODES IN GRID
CONNECTED TO A STORAGE CLOUD
S K Manu (1ks12cs083) Under the guidance of Murali Krishna V (1ks11cs048) Swathi.KNaseeruddin V N (1ks11cs053) Asst.ProfNavneet Kumar (1ks11cs055)
K.S Institute Of Technology Bangalore-62Department Of Computer Science
Batch-10
2
Problem Statement• To Reduce the computation time to process data by distributing the
processing task onto systems and providing high availability of data by using private storage cloud.
3
The architecture of storage cloud
4
Design
4
STORAGE CLOUD
Storage node
1
Storage node
2
Hadoop Master
Hadoop slave 1
Hadoop Slave 2
1 3
6
445
22
7
User
Controller node Of1 gb 1 gb
500 mb
500 mb
5
Result
5
Implementation Of Modules• User Interface• Openstack Storage Cloud• Hadoop Distribution And Processing
6
Testing And Results• Successful upload of data • Successful download of data• Successful map and reduce
7
Web page login
8
Simple User Interface
9
Upload Successful
10
Download Sucessful
11
Hadoop Map Reduce Process
12
Demo
13
Possible Outcomes• To process the data according to the user requirements.• High data availability.• Easy user interface.• Robust environment.• Reduced computation time.
14
References• 2015 IEEE International Conference on Computational
Intelligence & Communication Technology on Handling Big Data Efficiently by using MapReduce Technique.• http://docs.openstack.org/juno/install-guide/install/apt/
content/• http://stackoverflow.com/• https://www.google.co.in/?gfe_rd=cr&ei=HHoWVpqdD
uLI8AffhYH4DA
15
THANK YOU
top related