Designing a Customized Test Data Generator for Effective Testing of a Large Database

Business houses are investing heavily on new generation databases so as to gain competitive edge. As with any development project, a data warehouse also needs testing. The complexity and size of data warehouse systems make comprehensive testing both "more difficult and more necessary". The fact, queries that perform satisfactorily on small datasets may fail miserably in the real life environment. This necessitates establishing a system that runs queries on fully scaled data. Legal implications and business ethics do not allow to perform testing with real business data. In the paper authors have put efforts to generate test data ensuring correct balance and skewness making sure that the ratio of fact to dimension is correct and so on.