Research Of Multi-Dimension Query Optimization Algorithm Based On MDX

Recently, data warehouse becomes one of the most active branch ofthe database research field. And optimization of the OLAP query is thecore of this field. MDX (Multi-dimensional eXpression) proposed byMicrosoft, which becomes the de facto standard for many products, isexactly powerful tool to deal with compounded and multi-dimensionalqueries.The goal of multi-dimensional query and optimization is to makefull use of system's resource. Up to now, domestic and internationalscholars did many research works to deal with the problem of multiplequery optimization in OLAP environments, and proposed several ofalgorithms. But, in the past, the particularity of MDX query is notconsidered when they decomposed an MDX query into many subqueries.An MDX query often concern with many dimensional hierarchy. So, Annew MDX query model should be proposed.This paper studies the method of dividing subtasks and analyze theexisting algorithms, and has proposed multiple query algorithm based onsubquery divided. Main works include:1) We have compared the characteristics of SQL and MDX, studythree simultaneous optimization methods, and propose the analyse modelof execution cost.2) We analyze the existing multiple dimensional query algorithmsand study result caching technology, then provide the experiment data.3) We provide system structure of MDX query, and bring up an newmethod to divide the subquery, lastly design an new algorithm.Finally, the conclusion of this paper has been given and thedevelopment trend of the MDX query has been predicted.