Considering the group of mobile computing applications with occasionally connected clients and are wish to share data, nevertheless, due to the cost of mobile communication, they only connect occasionally. Usually a constant server connection, having a collective of client data, makes easy sharing between clients by permitting the clients to download updates submitted by other clients and upload local updates. Then the server computes and retransmits these updates on a client-by-client basis. As a result, limiting scalability because the difficulty of these operations is on the order of the number of clients. In this paper, focusing on developing a group basis approach that notably improves the scalability of the resources to synchronize data. Since synchronization is a problem with occasionally connected mobile databases. A server from time to time generates update files for each client, which are downloaded and applied when convenient. Miserably, the time required to synchronize clients in this approach increases notably with increase in client population. We show that this tendency could be changed by by grouping, update operation sharing is computed only once per group, irrespective of the number of clients. As the mobile devices will have limited memory capabilities we effectively utilize the memory of mobile device to store the data without using any backend software. When the mobile database is connected to the server it will have less session time. Therefore by taking session time into consideration the connection establishment and data transfer is to be done.