Discussion on some technical problems of the hotte

Discussion on some technical problems in GIS accessing network database

introduction in recent years, network technology has developed rapidly, which provides technical possibilities for the sharing of information resources As an information intensive geographic information system (GIS), it is at the right time to rise to the network platform However, from the current application situation, in addition to a few foreign companies having web version of GIS, it is still in the stage of experimental development in China Therefore, it is an urgent task for us to develop the prototype and products of network GIS with independent copyright in China as soon as possible and reach the international advanced level in technical means Several modes and characteristics of network computing (1) traditional centralized This is a host terminal mode. All computing tasks and data management tasks are concentrated on the host. The terminal is only an extension of the host input/output device The advantage of this mode is that it is easy to manage, but the disadvantage is that it requires high performance of the host, and also wastes the computing power of the computer as the terminal. From the perspective of performance price ratio, when the purchase cost is equal, the performance of a host is often not as good as that of the network composed of several computers; Therefore, this model has gradually withdrawn from the mainstream (2) Client/server (c/s) mode Generally speaking, in this mode, the server only manages the working principle of the digital impact testing machine, and the computing tasks are scattered on the client. The client and the server communicate through this is a very key technical network protocol of the tensile testing machine The client sends a data request to the server, and the server transmits the data to the client for calculation. After calculation, the calculation result can be returned to the server The advantages of this mode make full use of the performance of the client and greatly improve the computing power; In addition, since the communication between the client and the server is carried out through the network protocol, which is a logical connection, it is physically easy to expand at both ends of the client and the server It is the mainstream network computing mode at present (3) Browser/server mode In this mode, the client only needs a general browser, such as Netscape different step e or explore of two cylinders, which replaces all kinds of application software The server is a web server The browser and server are connected through tcp/ip The browser sends a data request, the web server takes the data from the background and calculates it, and returns the calculation result to the browser The advantages of this mode are: because the software used by the client is only a simple browser, users basically do not need training, and the client software does not need maintenance; Software upgrading and modification are only carried out on the server side, which is transparent to users; The server and browser can be on different operating system platforms Its disadvantages are: the web dynamic technology is not mature enough, and various standards need to be unified, such as the dynamic protocols released by various manufacturers do not support each other, browser disputes, etc In short, it is an advanced but immature technology Based on the above analysis, the client/server mode should be selected as the implementation mode of GIS accessing network database 2. The structural design of GIS access network database under c/s mode is generally divided into c/s two layers (see Figure 1), so as to make full use of the advantages of c/s mode, such as cross platform, easy expansion, data independence and so on The client side is designed in two layers - GIS function layer and data request layer. GIS function layer is the function implementation part of GIS, and data request layer is the data implementation part of GIS As an intermediate layer, the data request layer plays the role of data conversion. The upper layer is a data file with GIS characteristics, and the lower layer is a standard database record On the one hand, this form of hierarchical design makes full use of the existing single version GIS research results; On the other hand, the development of GIS function layer and data request layer can be carried out at the same time. As long as the interface standard remains unchanged, the change of this layer will not affect the other layer Fig.1 the general framework of GIS accessing database based on c/s model

it is worth mentioning the design scheme of ESRI's spatial database engine (SDE for short) (see Figure 2) It is the world's leading network computing model for GIS data processing Its data access form is: the user application requests spatial data from the SDE server through the SDE API. The SDE server stores the spatial object model, completes the spatial data search locally according to the characteristics of the spatial object, and returns the search results to the user's application through the network

comparing Figure 1 and Figure 2, we can see that both of them adopt c/s mode, and both of them deal with GIS function realization and data request hierarchically; The difference is the location of database oriented data request implementation: Figure 1 is implemented on the client side, and Figure 2 is implemented on the server side The main advantages of the implementation on the server side are: (1) the upgrading of the model and related computing mode for the output and consumption of spatial objects ranking first in the world can be implemented only on the server side and transparent to the client side; (2) Because SDE server and Oracle 7.2 database are closely combined, the data search speed is very fast For Figure 1, putting the data request layer on the client side, the degree of dependence on the database is different from that on the SDE server, which has a strong dependence on the selection of the database (currently SDE server is only implemented in Oracle 7.2). On the contrary, it is a very open structure, and the server it supports can cross not only the database system platform, but also the operating system platform It can be said that the advantages and disadvantages of Figure 1 and Figure 2 correspond to each other 3 Comparison of database access methods several methods of accessing database based on program are as follows (1) Special database access tools Such as Power Builder, Delphi, etc., it tends to manage and display the data in the database, and has limited computing functions It is not suitable to use it to develop GIS application system, and it is also difficult to closely combine their data manipulation function with the existing GIS application system (2) A conventional language embedded in a database language In order to allow user programs to directly access their own databases, database manufacturers basically provide proprietary C-Oriented precompiled headers and static libraries, such as open client of Sybase company and pro*c of Oracle. (3)