Main topic: How to promote data warehouse? Author: hem belongs forum: Data Warehouse question Points: 300 times Re: 1 Popularity: 6 Time: 2001-9-13 16:38:32 A data warehouse is a subject-oriented, integrated, time-varying non- Data sets for support management of decision analysis. The data warehouse analyzes the data of history to give the development trend of the analysts to support decision-making analysis. Used for online analysis processing. Database ordinary relational databases are mostly used in online transaction processing, handling current transactions, transactions. Examples of integration: Take a department store, there are many branches, each branch has its own daily transaction record data. There is a real detailed transaction history, and there is also a month, and the annual data is owned. The head office needs to understand the company's operations, to decide how to expand business, understand customer shopping trends, and promote new products and other series of decisions. Think about it, if each subsidiary sent all historical data to the head office, how many resources need to be stored? How much resources need to store these data? What do the head office do? The company does not need these detailed data. If the subsidiary is sent to the head office according to a certain cycle, so that the amount of data is small, it is conducive to the analysis. However, so many subsidiaries, each company's data structure definition may not be the same. Need a unified data definition. The data sent to the main firm is more comprehensive. In this way, the company can make a more comprehensive analysis of the subject of user shopping trends, shopping capabilities to explore customers and expand the company's scale. Topic: Communication companies may just record users' receivables, prepayments, call records, rates. If you create a customer topic area, you are used to record the services purchased by the user, and the user is located. This theme area can help communications to predict customer growth rates, customer's geographical distribution, customer purchase services to increase new services, and sell what communication products such as communications products in those areas. Change over time: This is mentioned above. The bookkeeping system of department stores, the communication company's accounting system is the best example of time, and the accounting system regularly belongs to the total data, and then the total data will continue to be added to the data warehouse, so that the data warehouse is also Change over time. Non-volatile: General, the data added to the data warehouse is not allowed to change, it is different from the online transaction system, allowing to modify the database record. This data in this data warehouse is generally not lost. There are several points in the data warehouse:
Integrated (extraction, conversion, loading) ETL
Metadata (data for describing data) Metadata
Theme star structure
The query in the data warehouse is the most important. View, index is to improve the selection of the query.
The code base is a unified, integrated premise.
