Improved Performance of Clusters using Job Grouping

Cluster computing is best characterized as the integration of a number of off-the-shelf commodity computers and resources integrated through hardware, networks, and software to behave as a single computer. The processes that are running on different processors on clusters may be in need of communicated with each other. If the processes that are going to communicate are not allotted to a single processor may suffer some performance penalties due to the communication overhead. A new scheduling algorithm is proposed using proper job grouping which is mainly based on the communication cost between the processes and are allotted to the same processor in the clusters and thus the performance is tuned such that the time taken for communication overhead is avoided. The experimental setup is done using Java and the results are analyzed which proves the effectiveness of the algorithm.