Performance Enhancement of Data Warehouse Using Minimization of Query Processing Proposal to Improve ROI

ABSTRACT The proposed work based on Pseudo Mesh Schema signifies that instead of a Centralized Data Warehouse, all the data will be resided over the geographically distant Work Station or Site, which is known as Local Site in the present work and among all the local sites, one of the Site will be elected as the coordinator Site, which does not contain any data at all, rather it contains a database consisting of MAC IDs of all the Decision Makers System which is attached to it. The proposed work also describes that the Users or the Decision Makers communicate with their coordinator only through a query and the Coordinator decides whether the request is from an Authenticated User or not, by searching the MAC_IDs stored in its own Database. Once the request from an Authenticated User is received by the coordinator, the coordinator starts to distribute or replicate the requested query over the Local Sites and waits for the response from the Local Site. It is to be kept in mind that, as the Dimension Tables are distributed over the Local Sites, all the Sites must not contain the same data. Therefore, the Site which contains the desire data of the query, responses first to its coordinator and then the coordinator sends a Stop Signal to all other Sites to stop their execution, and finally the Coordinator dispatches the result of the query to the Destination User. In this way the proposed work ensures the performance enhancement of Data Warehouse than the existing Centralized Query Processing Approach.