V. Distributed database development instance (refer to the SIPM system information developed by CIM, Shanghai Jiaotong University)
SIPM is an integrated CAPP system for process designers and process management, with powerful process design, process design process management, process sign judging, process version management, management process file automatic generation function.
Since we mainly introduce the distributed database, the information transmission process is discussed.
It features the data processing, the mass-oriented release information is triggered by the process task progress monitor, once the process is released and the network transmission license, all process information will be transmitted once, and timely information needs to be exchanged, due to information Store in the dispersion of each site (not only physical but also logically), it is very difficult to adopt active excitation applications, so the public data area passive method, that is, open a public data area on a dispersed database. The information that needs to be exchanged in a timely exchange of information is dynamically updated, and the application dynamically scans the update of each public data area. Once there is a new data, it is found to classify and update the status of the public data area.
In order to achieve the above objectives, the client / server-based computing environment distributed database system is used. The clients of each site are requested by remote procedure call (RPC) in SQL form to provide service providing services, and the server executes the required processing, The result is then returned to the client, the client, and the server are seamlessly computationally calculated by local area network; in the entire system, TCP / IP protocol communication is used between the local area network, and ensure that the data replication technology, two-stage submit agreement, etc. Consistency, integrity, and availability of group or heterogeneous data distributed on various venues of the network.
In order to facilitate the expansion and data exchange of the application system, a hierarchical order control type information integration method can be used, see Figure 5. Structure BOM, craft BOM, enterprise basic data Sub-system exchange data storage in the corporate center database (information in the enterprise center database is reinited, more integrity); information exchange between each subsystem passes data access The control interface is implemented to the enterprise center database; each subsystem can reappear information exchange through the hierarchical order control structure. The data access interface can be implemented by program control or through data attributes (such as validity control can be implemented by program comparison files, or can add timestamp fields in the database). This approach reduces the complexity of the interface between each subsystem, enhances the expansion of the entire integrated scheme; the exchange of data is concentrated between the various subsystems and the central database, improves the reliability of system integration; data exchange The function is performed between the enterprise center database and each subsystem, and the server center database is located, the calculation time required to share most of the data exchange can reduce the passive response of the subsystem exchange data exchange.
Figure 5
Sixth, conclusions
After introducing distributed database technologies, the contradiction between data dispersion and centralized management is effectively solved, and the sharing and exchange of data is realized. It turns out that distributed technology has an irreplaceable role in remote data management, and its prospect is more and more optimistic.
