Distribute Projects Across Nodes in a Cluster

You can distribute projects across nodes of a cluster in any clustered configuration. Each node can host a different set of projects, which means only a subset of projects needs to be loaded on an Intelligence Server. This provides you with flexibility in using your resources and better scalability and performance.

To distribute projects across the cluster, you manually assign the projects to specific nodes in the cluster. Once a project has been assigned to a node, it is available for use.

If you do not assign a project to a node, the project remains unloaded and users cannot use it. You must then manually load the project for it to be available. To manually load a project, right-click the project in the Project Monitor and select Load.

If you are using single instance session logging in Enterprise Manager with clustered Intelligence Servers, the single instance session logging project must be loaded onto all the clustered Intelligence Servers. Failure to load this project on all servers at startup results in a loss of session statistics for any Intelligence Server onto which the project is not loaded at startup. For more information, see MicroStrategy Community Knowledge Base article KB14591. For detailed information about session logging in Enterprise Manager, see the Enterprise Manager Help .

1. In Developer, from the Administration menu, point to Projects, then select Select Projects. Intelligence Server Configuration Editor opens, at the Projects: General category.

2. One column is displayed for each node in the cluster that is detected at the time the Intelligence Server Configuration Editor opens. Select the corresponding check box to configure the system to load a project on a node. A selected box at the intersection of a project row and a node column signifies that the project is to be loaded at startup on that node.

   If no check boxes are selected for a project, the project is not loaded on any node at startup. Likewise, if no check boxes are selected for a node, no projects are loaded on that node at startup.

   If you are using single instance session logging with Enterprise Manager, the single instance session logging project must be loaded onto all the clustered Intelligence Servers at startup. Failure to load this project on all servers at startup results in a loss of session statistics for any Intelligence Server onto which the project is not loaded at startup. For steps on implementing single instance session logging, see the Enterprise Manager Help. For more information about this issue, see MicroStrategy Tech Note TN14591.

   or

   If the All Servers checkbox is selected for a project, all nodes in the cluster load this project at startup. All individual node check boxes are also selected automatically. When you add a new node to the cluster, any projects set to load on All Servers automatically load on the new node.

   If you select a check mark for a project to be loaded on every node but you do not select the All Servers check box, the system loads the project on the selected nodes. When a new node is added to the cluster, this project is not automatically loaded on that new node.

3. Select Show selected projects only to display only those projects that have been assigned to be loaded on a node. For display purposes it filters out projects that are not loaded on any node in the cluster.
4. Select Apply startup configuration on save to allow your changes to be reflected immediately across the cluster. If this check box is cleared, any changes are saved when you click OK, but they do not take effect until Intelligence Server is restarted.
5. Click OK.

If you do not see the projects you want to load displayed in the Intelligence Server Configuration Editor, you must configure Intelligence Server to use a server definition that points to the metadata containing the project. Use the MicroStrategy Configuration Wizard to configure this. For details, see the Installation and Configuration Help.

It is possible that not all projects in the metadata are registered and listed in the server definition when the Intelligence Server Configuration Editor opens. This can occur if a project is created or duplicated in a two-tier (direct connection) project source that points to the same metadata as that being used by Intelligence Server while it is running. Creating, duplicating, or deleting a project in two-tier while a server is started against the same metadata is not recommended.