Install a Fusion 4.x Cluster
To support enterprise search applications at any scale, install Fusion on multiple nodes (a Fusion cluster). Deployed in this manner, Fusion can store large amounts of data and achieve high processing throughput.
| Historically, "node" has also had a meaning at Lucidworks related to licensing and to the number of CPU cores. We do not mean that here. In the context of this documentation "node" is a computer, usually a physical server but also possibly a virtual machine (especially for test environments). |
See also this video tutorial for building a multi-mode Fusion cluster:
To install Fusion on a single node, see Install Fusion on a Single Node.
| Failures in the Fusion install or startup may occur if the Fusion installation directory name contains a space. |
Install Fusion Server with Bundled Solr and ZooKeeper
When you install Fusion Server, the software includes bundled versions of Solr and ZooKeeper. Install Fusion Server on every node. Then, configure Fusion Server so that it does not start Solr and ZooKeeper on nodes where they are not needed.
Supported Cluster Arrangements
To satisfy processing requirements, install Fusion Server, ZooKeeper, and Solr on specific nodes.
Before proceeding, review the supported cluster arrangements.
Hardware and software requirements
Verify that the nodes on which you plan to install Fusion meet hardware and software requirements.
Next steps
Perform the installation steps in one of these articles:
If you already have a Solr deployment, use the instructions Integrate Fusion with an Existing Solr Deployment.