Efficient structures for geometric data management
Efficient structures for geometric data management
The R*-tree: an efficient and robust access method for points and rectangles
SIGMOD '90 Proceedings of the 1990 ACM SIGMOD international conference on Management of data
Research issues in spatial databases
ACM SIGMOD Record - Directions for future database research & development
Multidimensional access methods
ACM Computing Surveys (CSUR)
The Grid File: An Adaptable, Symmetric Multikey File Structure
ACM Transactions on Database Systems (TODS)
An introduction to spatial database systems
The VLDB Journal — The International Journal on Very Large Data Bases - Spatial Database Systems
Spatial Internet Marketplaces: A Grand Challenge?
SSD '97 Proceedings of the 5th International Symposium on Advances in Spatial Databases
| Hi-index | 0.00 |
As computing environments for Internet has been expanding, researches on Internet GIS (Geographic Information System) have been actively made. The omnipresence and easiness of Web make the research being focused on Web GIS. Web GIS is classified into Server-side Web GIS and Client-side Web GIS. This classification is based on the query processing location, the server or the client, respectively.The former can easily be implemented but raise the overload of the server since every request must be returned to the server and processed in server. The latter lessens the overload of the server, but raises some problems. Firstly, if client is not very powerful, large and complex datasets can be hard to be processed on the client. Secondly, the large amounts of data as well as applets or controls cause delays since they must be transferred from server to client.To solve these problems, this paper proposes a Jigsaw query processing technique that processes user queries dynamically. The proposed technique classifies the query into client class query, server class query or client-server class query. Client class query is processed in client itself. Server class query is transferred to the server and processed in the server and the result returns into the client. Client-server class query is processed as follows. The data required to process query is transferred from the server into the client. Next, the client processes the query with its original data and the transferred data. The proposed technique uses some meta data on client-stored data (such as spatial data, spatial index, and so on) and the communication speed between the client and the server to classify the class of the query, and uses tile-divided data structure for efficient spatial data transmission.For an evaluation, the detailed experiments such as map browsing, layer controlling, point query, and region query are conducted comparing the proposed technique with two other existent methods. Based on the result of experiments, the proposed technique can give the user fast response and increase system performance.