Clustering is the connection of multiple devices together in such a way that to the outside world, the cluster appears to be a single system. Each device in a cluster is referred to as a node.
If critical service is unavailable, business will suffer. A properly installed and configured cluster provides continued operation during hardware and software failure, and/or planned maintenance. Ability to recover from failed server hardware and failed software is the main benefit of clustering.
Windows Server 2003 clustering consists of two different technologies: Server Cluster and Network Load Balancing (NLB). Each of these technologies can be used to provide high availability for different types of services.
Server Cluster is primarily used to provide availability for mission critical applications through fail-over. It is typically used for databases, e-mail services and custom applications. Windows 2003 supports server clusters of up to eight nodes. Most clustered applications, and their associated resources, are assigned to one cluster node at a time. Node failure in a cluster is detected by the absence of heartbeats from that node. If a node fails to transmit a heartbeat packet for a designated period of time, that node is considered to have failed and the remaining nodes will takeover the work load of the failed node. In this case, the clustered application will be started on remaining nodes. Client requests are immediately redirected to the backup cluster node and the impact of the failure will be minimized.
