How can I improve the overall performance of MapViewer?
The limiting factors of MapViewer performance include: cpu speed/JVM speed, JDBC speed (fetching large amounts of data from db), amount of heap memory and spatial cache allocated to MapViewer. Always make sure you have allocated enough memory when starting the oc4j Java process. For instance, the JVM option -Xmx512M will ensure that MapViewer can obtain as much as 512 MB of heap memory when needed. Large amounts of memory are necessary if you will be rendering lots of data for certain map requests, or if you have allocated many renderers for each datasource defined. MapViewer scales well in a multi-processor environment. On such a system MapViewer’s capability to handle concurrent map requests will be dramatically improved, automatically. You can also set up a cluster of MapViewer instances to handle big workloads. Another option is to pin frequently accessed predefined themes entirely in the memory. This can be achieved by setting the caching mode of a predefined theme to “ALL” in the
