With Linux, it is a relatively easy job to configure multiple servers to run collectively as a ‘high-availability’ clustered system. The scalability of the system in terms of processing power and storage is potentially unlimited, and only constrained by budget. Continuity of service should one or more servers fail provides the ultimate in systems redundancy. The cluster is configured to continue operation even in the event of single or multiple server failure. Data is protected by building-in suitable redundancy to the storage network, as well as deployment of the ultimate failsafe of an off-site, remote backup regime.
Blade server configurations are ideal for hosting high availability clusters. They can be found in larger businesses or at service providers where many users rely on the servers to carry out their daily work, and where the combined power and reliability of multiple servers is needed for business critical functions. Where ultra-reliability and redundancy in case of failure are required, blade servers can be clustered together to provide a high level of availability should one or more of the individual blade servers ever fail. Servers take the form of blades and slide in to the slots of a master chassis. This solution is most commonly deployed with a fibre channel SAN for maximum consolidation and reliability.