Mediator Based Architecture is an emerging technology used to address data heterogeneity issues. As the number of data sources increases, the data integration process becomes an administrative and performance bottleneck because of data heterogeneity. Mediator-based approach allows the integration of data from heterogeneous data sources, which are usually not centralized. A mediator system is set between a number of data sources and applications. Now-a-days, the searching and combination of information from distributed, autonomous and heterogeneous software systems can be considered as a challenge for the computer users. The mainstream of this work is to address the issues of data heterogeneity where data are located in various distributed system in diflerent format. Hence, multiple data accessing point is required to retrieve information from various distributed system. The mediator service allows single point of access, enabling the retrieving of appropriate information. To achieve this goal, the service dynamically integrates and customizes data from various data providers. This paper concentrates the investigation of data sources, interfaces and capabilities for mediators, and the design and implementation of a model- “Integration Mediation Kit”. The kit can think the entire source query in a scalable way.