Data Warehouse Construction

Data Warehouse is a comprehensive technology and solution based on data management and application. The successful implementation of data warehouses has a significant impact on the cultivation of a culture of knowledge sharing. Currently, the data warehouse-based decision support system is mainly used in banking and securities industries.

With the intensification of global competition, more and more enterprises recognize that correct and timely decisions are the key to the survival and development of enterprises. Therefore, it has become the consensus of most successful companies to make full use of modern information technology to automatically and quickly obtain useful decision-making information and provide fast and accurate decision-making support for enterprises.

The emergence of data warehouses has given enterprises a better impetus for development. The word "Data Warehouse" first appeared in Early 1990s and has become mature. According to the IDC survey, the average ROI of data warehouses is 401%. According to another survey, 500 of happy 85% enterprises have established or are building data warehouses. Like the Internet, data warehouses are becoming the fastest IT growth point.
　　
　　　　 I. purpose of building a data warehouse
　　
1. the fierce competition in the market and the complexity of the management process determine the survival and development of an enterprise, it is necessary to extract and analyze the historical data of customer relations, marketing, product engineering, investment analysis, and other aspects to find potential information that is valuable to the further development of the enterprise.
2. The data warehouse can effectively integrate internal and external data of an enterprise and provide data basis for decision-making at all layers of the enterprise.
3. The existing system of the enterprise cannot provide more decision-making information (although the enterprise has accumulated a large amount of data ).
4. Construct a systematic data storage environment to separate the large amount of data required for analysis and decision-making from the traditional operating environment, converts scattered and inconsistent operation data into integrated and unified information.
5. provides basic information sources and auxiliary tools for marketing and customer analysis.
6. You can analyze the profits and costs of products, departments, and institutions.
7. Standardized management processes, optimized business processing, and improved capital utilization.
　　　　 2. Data Warehouse Construction steps
The data warehouse architecture consists of three parts: data source and Data Source conversion/loading to form a new database and OLAP (Online Analytical Processing On-line Analytical Processing ).
The implementation process of the Data Warehouse can be divided into three phases: project planning, design, implementation, and maintenance adjustment of the data warehouse.
Starting from the data warehouse architecture and implementation process, the data warehouse construction can be divided into the following steps:
1. Clear objectives and overall arrangement
Based on the enterprise's development goals and market changes, we will use the strategic development perspective to create an information architecture solution to ensure that the company's business objectives are consistent with the required data.
2. unified planning and step-by-step implementation
Building and maintaining an enterprise data warehouse is a time-consuming and labor-intensive project. Therefore, we should first design an overall information architecture, develop a phased implementation plan, and then gradually implement it, focusing on the data centers or data transmission mechanisms required for highly important business events.
3. Construct a technical environment and establish a Support Platform
Establish a technical environment, and select hardware and software resources to implement the data warehouse, this includes the selection of development platforms, DBMS, network communication, development tools, terminal access tools, and service level objectives (availability, loading, maintenance, and query performance.
4. Create a model and select a tool.
With the construction of the data model, enterprises can obtain complete and clear description information from it. The data model provides unified standards for multiple application data sources of enterprises.
The model design requires the company's information staff to work closely with the business staff to plan an application model that has practical value for the company. This model must be intelligent, the ability to constantly adjust its own parameters based on the actual business development, and ultimately find the rules in the enterprise's operation process, so as to bring benefits to the enterprise.
There are many tools for building a data warehouse, such as modeling tools, data purification tools, data extraction tools, data warehouse management tools, Online Analytical Processing and data mining tools. The world's famous big companies, for example, IBM, Oracle, Sybase, CA, NCR, Informix, Microsoft, and SAS have all launched product tools and application solutions for building, maintaining, and applying data warehouses. Enterprises can choose from the perspective of their actual situation and user needs.
5. Strengthen Management and maintain well
Data Warehouse security issues cannot be ignored. The operating system and database security mechanisms are used to strengthen the management of data warehouse operation permissions. Back up relevant data in the data warehouse in a timely manner, and use RAID configuration to back up the data warehouse to improve the security and availability of the data warehouse.
　　
　　　　 Iii. Problems in data warehouse system construction in China
　　
The data warehouse technology has not been well developed in China for the following reasons:
　　
1. China's basic information equipment is relatively incomplete.
2. The company's sense of competition and service is not strong enough.
3. Data warehouse prices remain high.
4. Lack of management mechanisms.
A data warehouse is a data sharing system. People at different levels may obtain different information. However, Chinese enterprises have not yet established a management mechanism to promote data sharing, and there is no continuous management mechanism for human abilities, organization systems, and data quality, it is very difficult to establish a good data analysis on this basis.
5. Lack of technical talents.
The application of the data warehouse is a process of establishment. During the establishment process, a large number of technical support personnel are required. From the domestic situation, there is still a lack of talents who can fully implement the data warehouse solution, which restricts the development of the domestic data warehouse market.
6. data mining tools are not mature.
In addition to OLAP, a higher level of Data Warehouse is data mining. However, there is no major breakthrough in this field, and the data mining technology in the market is still unsatisfactory.
7. Insufficient data accumulation.
A large amount of historical data is required for online analysis and processing. However, in addition to a few industries such as China Telecom, securities, and banking, data accumulation is insufficient.
　　
Data Warehouse is a comprehensive technology and solution based on data management and application. The successful implementation of data warehouses has a significant impact on the cultivation of a culture of knowledge sharing. Currently, the data warehouse-based decision support system is mainly used in banking and securities industries. With the maturity of various technologies, data warehouse will be widely used in finance, insurance, railway, aviation, retail, food, telecommunications, postal, medical and other industries.