Design And Implementation of Novel Job Scheduler for Map reduce System in Hadoop Platform |
Author(s): |
| Mohammed Muzammil Mehdi , RNS Institute of Technology; Prakasha S, RNS Institute of technology |
Keywords: |
| MapReduce, Hadoop Distributed File System (HDFS), Job Tracker, Task Tracker, Name Node, Data Node |
Abstract |
|
Apache Hadoop is an open source group of tools which is widely adopted in industries and as well as in academia due to its processing of large data in parallel. MapReduce is the most important component of Hadoop which is a programmable framework for pulling a data in parallel across the number of clusters. When lots of users submit their job at the same time they compete for the same resources as a result the job response time an important factor system performance seriously disgraced. So there is a need of efficient scheduler for such MapReduce clusters. To overcome this problem an efficient scheduler is designed and implemented which make use of the knowledge of workload patterns to improve the response time. |
Other Details |
|
Paper ID: IJSRDV3I40220 Published in: Volume : 3, Issue : 4 Publication Date: 01/07/2015 Page(s): 204-206 |
Article Preview |
|
|
|
|
