By Simon Gregory
Data replication is the process of moving data across the network so as to ensure consistency between geographically dispersed resources.
Data replication is the process of moving data across the network so as to ensure consistency between geographically dispersed resources.Traditionally replication has gone hand in hand with the protection it affords against data loss or disaster, but today businesses have additional requirements including consolidation, compliance and the immediate recovery of their data and information.
Replication of data comes in many forms, at the volume, file system, application, storage or data level. Our philosophy is to replicate at the data level - the key benefit of this approach enables businesses to abstract both the applications and the underlying storage infrastructure. As replication is already integrated into the data protection or disaster recovery solution we remove the need for additional software or processes, which means businesses can change or add applications and the underlying storage whilst the process of replication remains the same.
The current growing trend for adopting integrated replication is for the purposes of providing consistent data protection solutions, helping achieve compliance requirements and for providing disaster recovery. Managing data remotely is hugely expensive, so the advantage of removing the process of managing that data from the remote site becomes highly palatable. If 80% of data sits remotely (according to IDC), surely the easiest way to manage this is from the core?
Data replication is at its heart a relatively simple function whereby data is moved from a source to a target. However, moving data for disaster recovery and the most common causes of data loss, such as user deletion, data corruption, and loss of hardware, requires an integrated approach.
If a user deletes a file and wants to get it back, how does data replication actually help? On its own, it doesn't. By integrating replication into data protection, the user simply locates the file to be recovered and hits the recovery button. Replication is the necessary process for moving the data but it needs to be working in conjunction with the recovery solution to deliver a tangible business benefit.
To consolidate data for compliance or corporate governance means a number of things, retaining that data for prescribed time periods and allowing secure access to that data. Clearly data replication alone doesn't meet either of these requirements, it is simply the mechanism for moving that data after which businesses can then apply access and holding policies.
Given these user case scenarios data replication on its own is not enough. Business demands a simple mechanism for the retrieval of files in the event of a loss, consolidation of data for consistency, business efficiency, and providing disaster recovery if it is to meet current and future regulatory requirements.
Additional consideration needs to given when replicating data to ensure that it is secure, typically using encryption technologies. By replicating at the data level, businesses can simply apply encryption based on the policy associated with the data without the requirement for third party security infrastructure.
Additionally businesses should seek the most efficient mechanism for moving data, which would typically involve techniques such as compression, and deduplication thereby reducing bandwidth, costs and time. Again applying these technologies at the data level provides business with the flexibility to add or modify applications and infrastructure whilst maintaining the required service delivery.
So given the above, perhaps replication should be considered as a mechanism rather than the answer too many of the business requirements discussed. The true value of replication is only ever derived when integrated into a process that meets your specific use case scenarios.
By Simon Gregory, Business Development Director, CommVault.