The Ad-Hoc Query System Based On Multi-Dimension Data Model

This article which takes the Kun Ming telecommunication Decision Support Systems as a background, utilizes relevant technology of commercial intelligence such as data warehouse and modeling , has carried on more overall, more deep analysis and studies in owing the fee theme , has designed and realized a set of Ad-hoc query based on multidimensional datum structure which takes owing the fee as theme and carries on the money owed data analysis and the decision-making for the Kun Ming telecommunication has provided the powerful support.This article first introduces the Ad-hoc query system's generation background, and briefly introduces the Ad-hoc query the related technical of Ad-hoc query and the domestic and foreign data warehouse merchant in the Ad-hoc query technology development situation, then introduces the Ad-hoc query concept and the significance, explains the relationship of Ad-hoc query and the commercial intelligence relations as well as the Ad-hoc query's stratification plane locates in the commercial intelligence, meanwhile conducts the research and the explanation to the Ad-hoc query core question , the inquiry way and the classified question. Then this article takes the Kun Ming telecommunication money owed subject Ad-hoc query system as an example, described the realization in detail which based on the multi-dimensional data model Ad-hoc query system. Multi-dimensional data model design including subject area establishment, size selection and problem analysis angle (dimension) the selection is an extremely important foundational work, the present paper has paid the massive time for this, by understanding the angle of the question which the user cared about and user's true demand, finally designed a dimension level structure clear multi-dimensional model which user satisfiedly, at the same time realized is quite simple,The Ad-hoc query system that this text designs is based on structure of the multidimensional data, adopts ROLAP mode in storing the way, sets up and inquires about the systematic frame through strong commercial intelligence tool reportNet package and Frame Work multidimensional datum structural modeling environment of cognos impromptu, utilizes sqlserver2000 convenient DTS tool and data warehouse ETL to instruct the theory to finish the organization of the real data and cleaning work at the same time . This text is based on structure of the multidimensional data inquiry system, optimizes at Ad-hoc query efficiency .finally, according to withdraw multidimensional data memory of model realize way and star model fact form and link form many form related characteristic produce one form to multidimensional datum model star type specially Gathering optimized to construct inquire algorithm , carries on time complexity and space analysis and actual test of complexity to algorithm ,this analysis with test result indicates method which in inquire to multidimensional data model efficiency algorithm excellent than traditional gathering impromptu, can raise and inquire impromptu on the basis of the structure of the multidimensional data that inquire about efficiency in systematic gathering to a large extend .