Author : S.Rajan 1
Date of Publication :31st December 2017
Abstract: Scalability depends upon the nature of the problem and the algorithm which gives the solution to the problem. Load balancing balances the load between different virtual machines so that performance of the cloud can be improved by removing the load disparity among the Virtual Machines. Load balancing works well for the scalable Application. An improperly Scaled application will behave in an unpredicted manner. The resources such as RAM, CPU, Disk etc., may increase or decrease depending upon the need of the application so the resources should be allocated previously. This can be achieved by effectively scaling the application. Scalable Application decreases the costs as the costs are calculated based on the usage of the resource. To achieve the best scalable application in the cloud there is a need for a programming model which is highly fault-tolerant for the distributed file system such as GFS or HDFS. In this paper, we analyze the selected problem and the solution is framed based on the Map Reduce Programming Model. The efficiency of the scalability is studied through Parallel efficiency calculation. Based on the Parallel efficiency value effective Scalable Map Reduce Model is proposed
Reference :
-
- Ragu Ramakrishnan and Johanes Gehrke, Database Management system, McGraw-Hill, 2003
- S. Ghemawat, H.Gobioff and S.T.Leung. Google File System. In Proceedings of the 19th ACM SOSP Conference, Dec 2003.
- Analysis of Research Data using Map Reduce Word Count Algorithm, Manisha Sahane, Sanjay Sirsat, Razaullah Khan, International Journal of Advanced Research in Computer and Communication Engineering, Vol. 4, Issue 5 , May 2015
- J.G.A.D.a.1.LK.Ashwin Kumar, Hone: "Scaling Down Hadoop on Shared-Memory Systems" in 39th international Conference on Very Large Data Bases, Trento, Italy, 2013.
- L. L.S.P.A.R.Z.V.A.D.Wang Lam, "Muppet: Map Reduce-Style Processing of Fast Data" in 38th International Conference on Very Large Data Bases, Istanbul, Turkey, 2012.
- Emad Soltani Nejad Mohammad Reza Majma “A Modem Method to Improve Efficiency of Hadoop and Map Reduce Cluster Using Software-Defined Networks Technology” in 25th Iranian Conference on Electrical Engineering (ICEE), 2017
- Anca Vulpe “Exploring Scalability in Pattern Finding in Galactic Structure using Map Reduce” in 16th IEEE/ACM International Symposium on Cluster, Cloud, and Grid Computing, 2016