Query Optimization In Data Warehouse And Index Technology Research

Currently, business intelligence market scale flourishes unconventionally. It is estimated that in 2007 the production value of global business intelligence market will reach 20 billion U.S. dollars, while business intelligence has kept rising wave of mergers and acquisitions between corporations. Business intelligence has been increasingly recognized by IT industry, which has three main supporting technologies : data warehousing , online analytical processing and data mining. In recent years, from the renowned international database conference, as VLDB, SIGMOD, ICDE, we can find that these three technologies are the hot issues discussed in computer theory and industry.In this paper, the discussions focus on the data warehousing and on line analytical processing technology, with a focus on the optimization of data warehouse technology. Data warehouse is a kind of architecture, which is used mainly for decision support system. It should be able to support efficient data cube computing technology, access and query processing technology. There are different characteristics between data warehouse and database, such as quantity of data, the physical organization of data, operation types which they service, hardware utilization patterns and so on. It is decided that we can not copy database optimization experience for the data warehouse optimized. The main technology for the optimization of the data warehouse has materialized data cube and indexing OLAP data. Materialized data cube is expected to adopt the idea of pre-computing, and index has adopted the strategy of space for time. The papers focused on the indexes technologies. First, it is discussed that the principle of indexing and the typical traditional database indexing techniques, such as B-tree indexes , hashing index; then index widely used in the data warehouse technology, such as bitmap indexing, join indexing and join bitmap indexing; we will discuss a new indexing technology hierarchical bit-map indexing at last. It can efficiently index data of the data warehouse, according to the level and classification of the dimension. It can also be applied for data mining, information retrieval and other areas; hierarchical bitmap indexing has a good future.