A node is a host upon which one or more components have been installed. A cluster consists of Unwired Platform components running on one or more nodes that work together as a single, continuously available, system in order to provide seamless application management, device management, and data access to users.
Each node on a cluster is a fully functional part of the unwired system. In a clustered environment, the nodes work together to provide increased availability and performance.
Hosts in a cluster should have similar processing, memory, and I/O capability to enable load balancing without significant performance degradation. Hosts in a data tier cluster should be more powerful than hosts in an Unwired Server cluster.
There are different node strategies you can employ, depending on the environment you are designing:
- Single-node installations for personal development or trial installations – A nonredundant architecture consisting of an Unwired Server and data tier installed on a single host.
- 2-node installations for enterprise development and testing – A primitive architecture without load balancing that may optionally use a relay server. The data tier (which includes the CDB, the messaging database, and the monitoring database) is on one host and the server tier (which includes Unwired Server and Sybase Control Center) is on another.
- 3-node Unwired Server cluster for entry level production environments – A simple redundant architecture with two server tier nodes, and a separate data tier host. For an example of this cluster, see Redundant Architecture Options.
- N+2-node Unwired Server clusters and data tier clusters for full-scale production environments – An optimally redundant architecture with any number of server tier nodes, supported by a relay server and a data tier cluster. For an example of this cluster, see Redundant Architecture Options.