Remote replication is an essential part of data protection, providing a backup in case the primary site fails. Data is simply copied to other locations that may be simple data storage servers or fully capable secondary backup systems that kick in if the primary system fails.

Traditionally, it is just application data that is backed up, but now it is possible to replicate entire virtual machines that act as application or Web servers. These virtual machines are software implementations of actual servers and contain all the data, applications and configurations that are used in hardware servers, but they exist digitally, which means they can be replicated, transferred and run on any capable hardware and can be booted up in seconds to replace primary virtual machines that fail. What this means is that a hacker can bring down a virtual Web server in the primary location and, seconds later, backup virtual machines boot up from the remote location to shoulder the load, and users may not even experience any downtime or feel that they have been transferred to a different server running from a different location.

There are two types of remote replication:

• Synchronous remote replication — Data is replicated to a secondary remote location just as the primary data is being created or changed. This is real-time replication, or as close to it as possible, which ensures that data backups are, at most, only a few minutes older than the source material.


• Asynchronous remote replication — This is replication done not in real time as data changes, but in predetermined regular intervals such as weekly, daily or even hourly.