When establishing your data protection plan, there are a number of criterial which need to be taken into consideration. Amongst the most important of these are your optimal desired recovery time, and your tolerance for both planned and unplanned downtime.
Although any well-implemented backup plan will allow you to recover your data, this recovery process might involve many steps such as: purchasing new hardware, loading and configuring the system image, loading the last full backup, loading the latest incremental backup, etc…
For some critical business systems, such as ecommerce systems that must process high volumes of transactions, a slow recovery time is simply not ideal.
This is where high-availability systems can come in handy. High availability comes in various different forms. One possible example would be server replication.
With server replication, your critical systems are copied over to a secondary datacenter, which may also monitor your primary production environment. Both of these systems are identical, and constantly synchronized. If the primary site goes down, users can be re-routed to the second site as a failover option.
This is usually done very quickly, so that interruption is kept to a minimum. Most users will not notice any difference.
The secondary site will act as a temporary datacenter until the primary servers are ready to be brought back into production. We’ve put together a short video which outlines a possible use-case scenario for server replication as a high-availability mechanism.