| In this doctoral thesis, the author first gives a concise introduction about earthquake data and its management in China, including construction of earthquake data sources, categories and characteristics of earthquake data, media used for earthquake data storage, development of seismic information network, and management and standardization of earthquake data. A detail comparison is given between database and file system, and between centralized and distributed database system when used for the management of earthquake data. The result stresses the necessity and technological advantage of developing a distributed seismic database system. Based on the practices of the author in recent years, methods for the interconnection of database systems are discussed in detail, and five methods are put forward: two-phase commit, data replication, database gateway, remote server, and application-level interconnection. These methods can be used singlely or jointly, depending on the actual .situation. A Distributed Seismic Database System using Sybase products has been successfullyimplemented. Eqrthquake data are replicated near real-time through WAN betweenCenter for Seismic Information in Beijing, Seismological Bureau of Shandong Provincein Jinan, and Seismological Bureau of Hebei Province in Shijiazhuang. The goal ofdistributed data management and services through WAN are archived in this system.An experiment was carried out for implementing transparent access to Precursor Observation Database in MS SQL Server from Sybase ASE. The products, environment, methods, and procedures for this experiment are introduced. ASE users can access tables in remote servers as if the tables were local. This method has advantages of location transparency, DBMS transparency, easy maintenance of data consistency, easy management of data, and single user sign-on. And this method can also be used to connect Sybase ASE with data sources managed by other DBMS, such as DB2, and thus has wide application.In order for end-users to take full advantage of the Distributed Seismic DatabaseSystem, an application suite is developed. This suite mainly includes two packagesof software: the Precursory Data Processing and Analysis Software (PowerPre), andthe Seismic Data Processing and Analysis Software (PowerSeis). They are based onclient/server model, and can retrieve data directly from database managed by Sybase Adaptive Server and perform tasks of data processing. A great number of methods fordata processing and analysis are included into these two applications. They can be divided into three modules respectively: data management, data processing, and graphical result display. They have consistent user interfaces, rich graphical functions, and are easy to use.A Web site is set up to provide dynamic data services since 1998. This web site uses web. pb and a middleware application to link together the web server and the database server. Users can commit dynamic queries on data managed by Sybase ASE through web browsers. Earthquake catalog data and precursory data are in service currently. The set up of this web site has facilitated the sharing of earthquake data and enlarged the coverage of data services.Another distributed database system using Microsoft SQL Server is introduced in this thesis. The Center for Earthquake Precursory Observation Network collects digital precursory observational data of the capital area through this system from databases in Beijing, Tianjin and Hebei through database replication. Software for data transmission was developed for use when there is something wrong with database replication. The main function of the software includes data transmission and display of data and figures. This system has been put into production operation for more than one year, and its performance is satisfying.In the appendices, the author reviews the basic concepts and latest development of database technology, and gives an overview of database related technologies, such as data warehouse, OLAP, and data mini...
|
PDF Full Text Request