Drilling data shows the features of a large number, various types and dispersing. To make drilling decisions, these features can not support the analysis of multidimensional rapidly and effectively. Based on the HADOOP architecture, this paper puts forward a method of using the Apache Kylin framework in the multidimensional analysis of drilling information. In a BIG DATA environment, the optimization of cube model precomputation and technical proposal of inquiry processing in multidimensional analysis is discussed, which provides an effective way for multidimensional analysis of oil drilling information.