Countless IT processes count on copies of data. Backup is an obvious example, but test/dev, analytics processing, reporting, and file auditing are others. Various studies have indicated that copy data can consume as much as 10X the space consumed by the production copy of data. The total cost of copy data though is more than just how much capacity the process consumes. When IT calculates all the factors involved in creating and maintaining copy data they see ample justification for a copy data management solution.
Dissecting the Copy Data Process
When a copy of data is needed to feed an auxiliary process like test-dev, backup or reporting, there are a series of steps that need to happen. First, the IT administrator needs to quiesce the application that counts on the production copy of data so that when the copy occurs, all parts of the dataset are on the storage media rather than in RAM. Then the actual copy needs to be made to a storage system that the auxiliary process can access. IT then needs to point the auxiliary process that needs that data, towards the copy. Then finally, the auxiliary process can perform its functions. While all these steps are occurring, production data is changing, and IT needs to refresh the auxiliary copy, which means starting the process all over again and replacing the original copy.
The costs of the above process are tremendous. First, there is a steady stream of continuous interruption to the production application and more than likely it must stay in the quiesced state until the administrator finishes creating the secondary copy, which means each refresh takes as long as creating the original copy. Also, each time the copy is made it more than likely needs to traverse the network to the secondary storage system, and again each copy is a full copy, so each refresh consumes just as much storage as the first copy. Then once the copy is complete, the auxiliary application needs to stop, to attach to the new copy.
The entire copy data impact then becomes exponential based on the number of copies that are required.
How Copy Data Management Helps
A copy data management solution significantly lowers the impact of the data copy process. First, it automatically interfaces with standard applications. Also, once the solution makes the copy, copy data management can refresh that copy by sending it only the changed information. A change only refresh of the secondary copy takes only a few seconds. The application returns to full production performance almost instantly. The network only has to transport just the changes, refreshing the secondary copy automatically. Copies of the same data needed by other processes are then fed from the single copy under management, removing the exponential copy data problem.
To learn more about copy data management, the types of copy data management solutions that are available and how Hitachi is leveraging the technology with its solutions, join Storage Switzerland and Hitachi for our on demand webinar, “Evolving Data Protection from Backup to Copy Data Management”, which also includes a live demonstration of a copy data management solution in action.