Reducing response time in a distributed data warehouse system

Various approaches have been developed for storage and retrieval of data in a distributed data warehouse system. Most of these techniques focus on materializing views derived from the warehouse for answering similar queries in future. However, when a new unfamiliar query is made by a user, the system computes the answer by accessing large volumes of data present at various sites. Thus, the response time becomes unacceptably long and the system performance degrades substantially. In this research, we have proposed a model to accelerate real-time queries in these systems. This approach is based on building a surrogate structure for the primary key attribute of each relation. The relations are vertically fragmented and stored attribute-wise along with the surrogate values. We have shown that this model greatly reduces the query response time in a distributed data warehouse system.