Dynamic heterogeneous data processing in IaaS cloud

The integration of framework in cloud computing companies makes the customers easy to access and deploy their programs. Currently used processing frameworks are static and homogeneous in nature. So the initially allocated computing resources may be inadequate for big parts of the submitted job which increases processing time and cost. Nephele, a first data processing framework utilize the dynamic resource allocation offered by IaaS clouds in task scheduling and execution. Different types of virtual machines can be assigned to a particular task of a processing job which is automatically instantiated and terminated during job execution. Nephele with Mapreduce concept is existed but Mapreduce perform well among homogeneous datasets. Mapreduce merge is the alternative to Mapreduce in processing heterogeneous data relations. In proposed system, Mapreduce merge in Nephele to process data relationships among heterogeneous datasets is illustrated. Based on this new framework extended evaluations of Map reduce inspired processing jobs on an IaaS cloud system is performed. Finally, the results are compared to Map reduce merge inspired jobs on an IaaS cloud system.