抽象的

Dynamic Load Rebalancing by Monitoring the Elastic Map Reduce Service in Cloud

Suriya Mary , Vairachilai

CLOUD Computing is a compelling technology. In clouds, Distributed File Systems (DFS) are sharing their resources to process the large amount of typical data. It uses imbalanced chunk relocation process using the centralized approach that causes adequate failure in real time operations. This results in load imbalance in the distributed file system. A fully distributed load rebalancing algorithm is presented to cope with the load imbalance problem. A computational cluster designed specifically for storing and analysing huge amount of unstructured data in the distributed computing environment. Finding the storage nodes capacity from the cluster formation can be able to resize the data nodes and task nodes depending upon the input files. Amazon CloudWatch provides monitoring for AWS cloud resources and its applications. Amazon EMR records metrics that can be used to monitor the cluster. The load balancing progress can be tracked by running the cluster. Amazon CloudWatch automatically read the metrics of the DataNodes and TaskNodes such as CPU Utilization, number of read and write operations performed on the disk. Rebalance the load of the DataNodes based upon the user privilege by using the monitored alerts dynamically. Hence the monitoring solution is reliable, scalable and flexible.