Utilizing automatic load balancing

Cluster environments in releases earlier than 12.4.5 required you to manually load balance the traffic between the report servers within the cluster. In release 12.4.5 we introduced automatic load balancing based on unique users. This solution however, was prone to balance the traffic load unevenly. Some unique users may contain different amounts of data associated with them for example, number of operations. In such instance, while the number of users is balanced evenly between all nodes, some nodes will have a larger processing load. This, combined with possible differences in system hardware, may cause for some the nodes to perform diversely.

To receive a valid configuration with the proper farm hierarchy, consolidate all single node clusters with the same configuration into one cluster. The clusters that will be converted into nodes will balance the workload and acquire their configuration and settings from the cluster primary node.

Important

In your legacy deployment, if you were using the client and server range settings or application and service names to load balance the incoming traffic, we recommend that you review these settings, as they may cause a portion of incoming traffic to be ignored.

To modify these settings in NAM 12.3 farm configuration:

  1. Log in to the primary node in the load balancing cluster and go to HTTP://<CAS_URL>/diagconsole .

  2. Under the Configuration Management section click Advanced Properties Editor.

  3. Using the search box, find rtmjob properties.

  4. Modify or clear the RtmJob.client, RtmJob.server , RtmJob.service_name or RtmJob.appl_name settings.

Consolidation of the nodes into one cluster involves removing a node from a single node cluster and adding that node to the load balancing cluster. This involves a step where you can select a database reset for the node that you are adding. If the new node has different site and user option settings from the primary node's settings of the edited cluster, that node will be automatically marked for a database reset in the Clear nodes database screen of the wizard. Since this action is required, the marked option will be disabled. Both, Site options and User options of the Central Analysis Server Configuration must be matching the primary node. If you are tracking user IP addresses from selected ranges and, these ranges differ between the primary node and the node being added, we recommend that you remove the User IP address ranges in the configuration options of the node that is being added. For more information, see CAS Configuration.

To reconfigure your cluster for automatic load balancing:

Remove all nodes from any existing clusters within your farm.

If you were using the client and server range settings or application and service names to load balance the incoming traffic, we recommend that you review these settings for each of the report servers that will participate in the new load balanced cluster, as they may cause a portion of incoming traffic to be ignored.

Recreate a farm creating load balanced cluster. For more information, see Configuring report server farm.

Optionally, you can recreate any necessary filtering.

Filtering

Keep in mind that the Primary Node will first load balance the traffic and then apply any filters that you might have set up. This may cause a portion of incoming traffic to be ignored.