As the era of big data is coming, it is more and more difficult for the traditional GIS analysis tools to meet the analysis needs of large volume spatial data. SuperMap iServer 9D provides the distributed cluster architecture of geographic information service. It can make spatial analysis for large volume spatial data based on cluster and the Spark distributed computing method. Then, how to build the distributed cluster environment? The following parts will illustrate the whole process based on SuperMap iServer 9D (built-in Spark).
Experiment environment description
First, let’s deploy SuperMap iServer both on computer A and computer B, and build a computing node distributed cluster system on each computer.
Deployment process
1. Configuring the master code
The operation process is as follows:
1) Launch SuperMap iServer.
2) Configure and launch the native Spark cluster.
3) Add a reporter and set the cluster service address of computer A as the report address. Select “Enable the reporter” and “Enable distributed analysis node”.
4) Enable the distributed analysis service. Then we can check the enabled cluster and the nodes which are added into the cluster in the cluster overview.
2. Configuring the sub node
The operation process is as follows:
1) Launch SuperMap iServer on computer B..
2) Configure and launch the native Spark cluster.
3) Add a reporter and set the cluster service address of computer A as the report address. Select “Enable the reporter” and “Enable distributed analysis node”. Please do not enable the distributed analysis service when configuring the subnode.
3. Check the Spark node
We can check whether the sub nodes have joined the cluster in the master node cluster overview. Also, we can check whether the Spark node has been configured successfully through the 8080 port of the master node. The quantity of Workers represents the quantity of Spark node.
4. Distributed analysis
After the 1-3 steps, the distributed cluster deployment has been completed. Then, we need to check whether the environment has been deployed successfully by using distributed analysis.
After that, the SuperMap iServer big data distributed cluster environment has been deployed successfully.
Satellite Tracking + Big Data GIS Application in Real-time Ship-controlling>
Big Data - Signaling Data Makes Cities Smarter>
