Data reduction is limiting the amount of data. This can be done on more than one place. In the network or on the storagesystem and takes place without any notification for the end user. The goal of data reduction is allways: REDUCTION OF COSTS.
Different kinds of data reduction:
To realize the reduction there are many different methods with specific properties. Data reduction can be realized in the (transport)network or on the storage hardware itself. The method of data reduction depends on the environment of network, the users and the place of the data.
Reduction during transport:
By optimization, cache and/or de-duplication it's possible to reduce the amount of data between one or more locations. Qi ict partners with a solution for transport data reduction are Riverbed and Blue Coat. Using data reduction can save costs on the bandwidth between the locations. Backup windows become smaller, bigger files can be used on the remote site in a compete centralized environment. Without data reduction this is not possible.
Reduction on the storage system:
Another solution is to compress and deduplicate the existing data on the storage hardware.
With this solution the existing data will be optimized inline without any notification to the user. By using this method, less storage is needed to store the same amount of data. You can save up storage hardware but also on your back up window, restore time and bandwidth to a disaster site. Qi ict partner Ocarina Networks uses this kind of technique. Off course the end user doesn’t notice this process and is able to access his original data without noticeable latency.