Resource management in cloud data centers

Abstract

Vast sums of big data is a consequence of the data from different diversity. Conventional data computational frameworks and platforms are incapable to compute complex big data sets and process it at a fast pace. Cloud data centers having massive virtual and physical resources and computing platforms can provide support to big data processing. In addition, most well-known framework, MapReduce in conjunction with cloud data centers provide a fundamental support to scale up and speed up the big data classification, investigation and processing of the huge volumes, massive and complex big data sets. Inappropriate handling of cloud data center resources will not yield significant results which will eventually leads to the overall system’s poor utilization. This research aims at analyzing and optimizing the number of compute nodes following MapReduce framework at computational resources in cloud data center by focusing upon the key issue of computational overhead due to inappropriate parameters selection and reducing overall execution time. The evaluation has been carried out experimentally by varying the number of compute nodes that is, map and reduce units. The results shows evidently that appropriate handling of compute nodes have a significant effect on the overall performance of the cloud data center in terms of total execution time.