Dissertation > Excellent graduate degree dissertation topics show

Research and Implementation of Data Cube Techniques Based on QC-Tree

Author: LiYiChen
Tutor: ZuoJianQing
School: South China University of Technology
Course: Computer Software and Theory
Keywords: Data cube QC-Tree Multidimensional query language OPC BFSC
CLC: TP311.13
Type: Master's thesis
Year: 2011
Downloads: 12
Quote: 0
Read: Download Dissertation

Abstract

Increasingly wide application of data warehouse, in the context of the rapid development of business intelligence and decision support technology, massive data processing capacity to a higher demand. As an important application of technology for data warehousing, online analytical processing (OLAP, Online Analysis Processing) technology for management decision-makers to provide multi-angle, fast, intuitive view with a great deal of analysis flexibility. The data cube is an important means of data warehouse, data organization and storage, only pre-computed data cubes can significantly reduce query response time, providing the performance of OLAP. However precomputed will lead to the explosion of the cubic volume of. Therefore, there is a need to explore the cube structure and query performance index can effectively improve the technology to solve these problems. The quotient cube (Quotient Cube) is a classic lossless compressed data cube storage structure, and it was first proposed the data cube semantic retention issues, but does not solve the problem of incremental updates. Use of the prefix shared ideas, QC-Tree is constructed on the basis of the quotient cube a semantic query tree, to further reduce the storage space and provides efficient query performance and incremental update. Found through experiments, the the traditional QC-Tree is divided into two phases of the construction algorithm efficiency is not high. This topic in-depth study of the QC-Tree structure the DFSC the characteristics and traditional construction algorithm based on an effective single-stage construction algorithm OPC generate a temporary class and constructed QC-Tree can be simultaneously to solve the traditional algorithm memory occupy much of a problem. Then based on the breadth-first construction algorithm BFSC, in order to further improve the QC-Tree structure speed. Finally, OLAP analysis server query, calculation and metadata definition and QC-Tree point query methods, multidimensional query language MDX (Multi-Dimensional Expression) parser, MDX statements extracted metadata to be answered the answer given by the QC-Tree group, the comparison of the results with the Microsoft SQL Server Analysis Services QC-Tree structure prove that the new algorithm to answer queries correctness. The experiments and analysis the OPC algorithm and BFSC of algorithm structure performance are better than the traditional construction algorithm from memory space, construction speed, query answering other aspects confirms the validity and correctness of the algorithm. Finally, from the QC-Tree storage space and query performance discussed further optimize the technology of the QC-Tree.

Related Dissertations

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Programming > Database theory and systems
© 2012 www.DissertationTopic.Net Mobile