Single-level cluster structure

Feedback

A single-level cluster structure is composed of one cluster service level and one GIS application level, as shown in figure 3. The cluster service level can have multiple cluster service nodes, and the GIS application level can have multiple GIS application nodes.

Figure 3 A single-level cluster structure

A cluster service node is responsible for the following three tasks:

  1. Monitors information on GIS application nodes: It monitors the node information of the lower level GIS application nodes. The information includes the address information, load information, and current status information of the GIS application nodes;

  2. Updates information on cluster service nodes: It updates the load balancing information of the cluster service node based on the obtained information on GIS application nodes, so that the tasks requested by the client can be accurately allocated based on load balancing information;

  3. Sends feedback to the client: It receives the service request from the client, and then selects the GIS application nodes that satisfy the specified service request based on the load balancing information of the cluster. The node information of the selected GIS application nodes is returned to the client so that the client can be automatically connected to the GIS service that is best for its request.

A GIS application node is responsible for the following tasks:

  1. Reports information on the GIS application node: It reports its current operating status to the cluster service node. The reported information includes information about the GIS services provided by this GIS application node, the load information, and the current status information;

  2. Provides GIS services: After the cluster service connects the client and the GIS application node, the GIS service performs GIS calculation based on the GIS service request submitted by the client. The calculation result is then returned to the client.

  3. Receives/Sends service requests: If this GIS application node can be directly accessed from the client, such as GIS application node *1 in figure 1, the node can directly process the received request from the client. Additionally, it can send the service request from the client to the connected cluster system. The cluster system then selects a GIS application node that satisfies the service request based on the load balancing information of this cluster service. The node information of the selected service node is then returned so that the client can be automatically connected to the GIS service node that is best for the request.

  4. Receives information: It is also used to receive feedback from the cluster service about the best GIS service that satisfies the specified service request. The received feedback includes address information, load information, and current status information. All the information is returned to the client.