Using Distributed Hash Table Balance the Loads of Nodes

  IJCTT-book-cover
 
International Journal of Computer Trends and Technology (IJCTT)          
 
© 2015 by IJCTT Journal
Volume-27 Number-3
Year of Publication : 2015
Authors : S.naziya
  10.14445/22312803/IJCTT-V27P125

MLA

S.naziya "Using Distributed Hash Table Balance the Loads of Nodes". International Journal of Computer Trends and Technology (IJCTT) V27(3):144-147, September 2015. ISSN:2231-2803. www.ijcttjournal.org. Published by Seventh Sense Research Group.

Abstract -
Distributed file systems are specific for large scale systems and energetic in clouds. These DFS are input structure blocks for cloud applications based on the MapReduce programming paradigm. In DFS each file node is portioned into number of chunks, these chunk nodes are fixed sized nodes in HDFS and bill the distinct node that’s way MapReduce task can be perform in equivalent nodes. In cloud node failure is a norm. Nodes may be added, deleted and replaced in the distributed file systems. In the past chunk nodes are not efficiently distributed in nature it strongly depends on central node for reallocation. This one not satisfactory in the largescale systems so concert is entirely block and movement cost is high. To overcome these problems we propose a load rebalancing (LR) algorithm and planned as a DHT. The chunk nodes achieve our load rebalancing sporadically balance their loads and reduce the movement cost. DHT is unique identifier for chunk servers, it reduce the latency.

References
1. Shenbaga Bharatha Priya, A., J. Ganesh, and Mareeswari M. Devi. "Dynamic Load Rebalancing Algorithm for Private Cloud", Applied Mechanics and Materials, 2014.
2. Beer, Wolfgang, Christian Derwein, and Sandor Herramhof. "Implementation of Context-Aware Item Recommendation through MapReduce Data Aggregation", Proceedings of International Conference on Advances in Mobile Computing & Multimedia - MoMM 13, 2013.

Keywords
Distributed File Systems, cloud computing, Load Balance, Clouds, Map Reduce.