Clustering servers to prevent down time
A common feature of load balancing is clustered servers. A clustered server can be defined loosely as any two or more servers working together in a coordinated fashion to deliver a service or set of services.
Considering the demands of Internet-based applications, IT organisations must keep the enterprise infrastructure constantly working at a consistent, responsive level of performance, especially during reasonable peaks in user load.
Clustered servers are designed to improve system reliability and manageability. IT organisations deploy failover clusters to keep end-user applications operational during both planned system events (such as backups, maintenance, and peak user loads) and unplanned events (such as hardware failures and network outages).
As part of the fault-tolerant design, each critical system component sends a short message, or heartbeat, over the network at regular intervals to indicate that it is still online.
If a critical component has failed cluster members will initiate emergency actions to assume the load and resources of the failed node.
Clustered servers have three main benefits:
- Increased availability
- Greater manageability
- Enhanced scalability
For more information on clustered servers click the link or request more information online.