Study On Performance Optimization Method Of SVG-based WebGIS

In recent years, GIS has entered a new Web era. The emergence of the WebGIS makesGIS more socialization and popularization. However, it still faces severe challenges.Currently, most of WebGIS solutions are based on the raster images as the visualexpression of spatial data. On the one hand raster images must be rendered by server, onthe other hand, the static raster images translated to the client can only be used to display,but can not be used to implement GIS functionalities. Server takes all GIS operations andbecomes overloaded, especially in multi-user concurrent access, the problem is especiallyprominent.The client browser support of SVG gives us an alternative common cross-platformvisual expression way for WebGIS. By transmitting XML-based SVG text, then we canalso achieve graphical representation of spatial data. More importantly, we can reuse theresult data. Give full play to the operability of the vector data will greatly enhance theperformance of WebGIS.Based on the above facts, the vector-based WebGIS has become a hot topic and hasmade certain finding. However, there are still many difficulties in the studies ofSVG-based visual expression of WebGIS. Especially when describe large amount of spatialdata, it will cause server bottlenecks and network traffic bottlenecks. This paper integrateswith the uniqueness of SVG, focuses on how to improve overall system performance andproposes a SVG-based WebGIS solution. Finally, we apply the solution to the approval ofconstruction land management to verify the feasibility.The main contents of this paper can be abstracted as followings:(1) A GZIP compression module is imported into the system between the processes ofvector conversion and result data transmission. In order to suppress strong influence on thedata amount of vector data results by the size of the spatial data.(2) To avoid long time waiting on GIS requests, we design and implement fragmentedtransmission of SVG data. It splits one heavy request into multiple light requests. Once alight request has been done, the result will be return back immediately.(3) We also try to simplify the spatial layers according to the different map zoomlevels and different tolerances. By customizing data dictionaries of layers to change theway of spatial data visiting. In order to adapt the situation of low zoom and large amountof data transmission. (4) We enhance the efficiency of SVG data conversion by pre-storing the vectorelements into spatial database tables. This can help us to reduce the process time of dataconversion. It is done in independent spatial objects granularity, so it can well adapt to thedynamic update of spatial database.For vector-based WebGIS, especially for SVG-based WebGIS, the researchachievements in this paper are helpful for performance improvement, enhancing theversatility and promoting practical application.