I. Introduction
Since the 1980s, database technology has been greatly developed, especially the emergence of distributed network database technology, so that data in different regions have been shared, and the coordination and efficiency of work have been improved.
In some companies that have geospadial subsidiaries. The dispersion of the geographic location has caused the dispersion of business data, and the head office is in different cities or the various regions in the city. In addition to handling their respective data, they also need to exchange and handle each other. How to deal with dispersed data and centralized management, which is a difficult problem that is sleepy around the database developers, and the emergence of distributed database system technology provides possible. With the development of computer network technology and geographically dispersed sectors, companies, manufacturers to extension, database technology extends from single machine to the network, collect, storage, processing, and propagation of data from a centralized distributed, from enclosure Opening is inevitable.
Second, distributed database technology introduction
The distributed database system is a client / server architecture, which is shown in Figure 1. In a network environment, each hardware platform with multi-user processing capabilities can become a server, or it can be a workstation. Databases on multiple servers Figure 1 Distributed database system structure is a logically single database overall, data consistency, integrity, and security are controlled to this logically single database. The server is managed for the access of shared data, and the processing operation of the Non-Database Management system can be done by the client.
user
The internet
DBMS
server
DBMS
server
DBMS
server
user
user
The combination of distributed (network) technology and database technology is logically belonging to the same system, but a set of data sets of multiple venues (nodes) that are physically dispersed in computer networks. Conceptually, distributed databases are physically dispersed on each node of computer networks, while logically belonging to data sets of the same system.
The distributed database has the data distribution and the two major characteristics of coordination between database. The system emphasizes the autonomy of nodes without emphasizing the centralized control of the system, and the system should maintain data distribution transparency, allowing the application to prepare the distribution of data at all.
The distributed database system has appropriate data redundancy by replication, but can increase the reliability and availability of the system; provide a coordination between local autonomous data sharing and venue, so that the system has fast data processing power; Based on database technology and parallel processing technology, the scale benefits of multiprocessor parallel processing can be used to improve the rapid response capability of the system.
The data on each site (node) is generally used to describe the real world of this site, the data source of the local database and most users (applications) are generally reside in this field, that is, each site has independent processing. (Site Autonomous), can perform local applications; in addition, global applications can also be performed through network communication through network communication. For the user, a distributed database is logically, just like a centralized database, the user can perform global applications in any site.
Data independence concepts in distributed database systems are equally important, but a new concept is added, which is distributed transparency. The so-called distributed transparency is that the data is not distributed when writing a program, so the transfer does not affect the correctness of the program.
