Abstract
In the recent years, there has been a substantial increase in data growth rate. Accordingly, a powerful framework for analysis and processing of big data is required. Hadoop and MapReduce Framework is a suitable framework for resolution of this problem. It eliminates complexities of parallel programming such as data distribution, timing, and node failure. Application of big data in Hadoop cluster network is characterized by large amount of eastern-western traffic, changes in traffic patterns in dynamic manners, and immediate nature, which have caused several challenges including traffic increase and congestion. In this article, a modern method to improve efficiency of Hadoop and MapReduce Framework is presented that is activated using Software-Defined Network (SDN) technology based on network traffic and operates combine function. OpenFlow protocol is tested on a Hadoop cluster with ten artificial nodes in Cloudera. Results of this test on high traffic load networks are expressive of improvement of Hadoop cluster efficiency using OpenFlow protocol.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
