Entwicklung eines Query Prozessors in einem Föderierten Data Warehouse System

Mergers of companies may require sales analysis across several Data Warehouses, where there is no central authority, which manages the data. For solving this problem a federated Data Warehouse System will be an appropriate approach. Due to the fact that the system provides a global schema, a federated Data Warehouse System gives the user transparent access to the underlying schemas and data. Moreover it is necessary to hide the heterogeneities, which may occur during the integration of Data Warehouse Systems. Thus the multidimensional language SQL-MDi was developed to resolve the heterogeneities. Goal of this thesis is the design and the prototypic development of a processing-component for SQL-MDi, which is part of a query tool in a federated Data Warehouse System. The principal task of the query tool is to generate a global, integrated cube. The main challenge of the Query Processor is to traverse and evaluate an operator-tree and to calculate a global result. The operator-tree is generated by the Query Parser from an SQL-MDi-Query. The tree-structure contains the operators of the Dimension- and Fact-Algebra on which SQL-MDi is based on. The operator-tree is used as an evaluation plan and can be traversed by the processing-component without any restructuring.