Data replication is the practice of creating one or more redundant copies of a database or other data store for the purpose of fault tolerance.
Data replication is not a static backup of data. Instead, data replication encompasses duplication of transactions on an ongoing basis, so that the replicate (often called a mirror) is in a consistently updated state and synchronized with the source. There are different methodologies of performing replication depending on the specific availability requirements for the data.
Problems with data replication can arise from latency or service interruptions during data transfer. Commercial data replication products attempt to alleviate business risk due to replication errors or failures. As the distance between the source and the mirror increases, data replication can become more challenging.
The primary benefits of data replication are disaster recovery and high availability of mission-critical applications. If the primary data source should fail, a replicate can be swapped in immediately. It also provides transactional consistency so that the data is up-to-date and consistent. Data replication tools can reduce the IT labor involved in creating and managing data replication transactions across the enterprise.