Heterogeneous big data processing poses significant challenges due to the diverse nature of data and the varying computational capabilities of processing resources. Map Reduce is a common programme paradigm for handling massive amounts of data, but its performance on heterogeneous environments is often suboptimal. This paper presents an integrated model that aims to enhance the performance of Map Reduce on heterogeneous big data processing. The model incorporates several techniques and optimizations to efficiently utilize the available resources and minimize the impact of resource heterogeneity. Experimental evaluations determine the efficiency and advantage of the suggested model in terms of performance improvement, resource utilization, and scalability. The results specify that the integrated model can significantly enhance the performance of Map Reduce on heterogeneous big data processing scenarios. In proposed system, designed a new scheduling algorithm, Speculating Prioritize Tasks (SPT) algorithm that is very resistant to diversity. In clustering of 200 virtual machines on Elastic Compute Cloud (EC2), SPT may enhance Hadoop speed of response through a factor of two.