Based On Relational Databases, Olap Strategy

On-Line Transaction Process (OLTP) has been incapable of meeting the needs which terminal consumers have about querying and analyzing of database, and it has also not been satisfied consumers that SQL queries large database. Decision analysis of consumer requires a large number of calculation on the big database, which can get the result, and the result of querying can not meet the needs of decision-maker yet. So the concepts of multidimensional database and multidimensional analysis are put forward, viz. On-Line Analytical Process, it gives analyst, manager, or executor the ability of information access which is rapid, accordant and interactive from multi point of view. The information comes from original data, and can be catch on by consumer and really reflects the character of enterprise dimension. So OLAP is one kind of software technology which can make them know better of data. ROLAP is common one of realization mode of OLAP and the realization of ROLAP is basically obeyed tri-tier Client/Server structure. After the data model of data warehouse is defined, the data which comes from all kind of different data source will be cased into data warehouse, and then system will run the corresponding program to integrate the data based on the data model and create indexes to optimize the access time. The requirement of multidimensional analysis of terminal consumers will be dynamically translated to SQL requirement by ROLAP engine, then relational database will handle the SQL requirement. In the end the requesting result which is handled by ROLAP engine will be sent to consumer as multidimensional cube.The article addresses one way to integrate the ROLAP technology and traditional report form function, and the way can manage the searches of the relational database to realize multidimensional data analysis and to deal with the execution ofreport form continuously. The multidimensional structure is the stanchion of decision support system and is also OLAP core. The article studies how to dynamically create Multi-Cube structure using SQL language and ROLAP engine and to realize multidimensional calculation combined with OLAP engine and to satisfy the report form and multidimensional analysis requirement of consumer by drive of user's analysis.