michelangelus - Fotolia
Most organizations that have been around for a while seem to struggle with managing and maintaining secondary data storage. It's large, unwieldy and, in most cases, siloed.
In fact, it is common for a large organization to have a dozen or more separate data silos.
Having data locked away in silos presents at least two major problems. First, it can be difficult to adequately protect this data. If data is locked away in a silo, then there is a good chance it is protected by something other than the organization's primary backup and recovery platform.
The other issue with having data silos is they essentially cause the data to be treated as single purpose in nature. While there may be a tangible benefit to allowing the data to be used by other systems within the organization, the way in which the data is stored can make it nearly impossible to use for any purpose beyond its original intention.
Disk, hyper-convergence support secondary data
Although it may seem somewhat unusual, some organizations have been able to dramatically improve the ability to put their secondary data storage to good use by changing their data protection strategy and using hyper-convergence and disk-based backups.
Disk-based backups have long been known for their ability to eliminate the backup window and to provide a far better recovery point objective than could realistically be achieved with a tape-based system. Many modern disk-based backup platforms even provide instant recovery capabilities for virtual machines. As impressive as these backup products may be, modern hyper-converged platforms take things a step further by helping to make previously siloed data available for use within the organization.
A hyper-converged backup product can bring all of an organization's data together into a single location. These systems can then use copy data management to make that data available for use throughout the organization.
Secondary data storage benefits from copy management
Copy data management is a technique for creating virtual copies of data. Imagine, for example, a team of developers needs a copy of a particular data set for testing purposes. The normal way of handling this would be to copy the data for the developers. This method isn't ideal, however, because the copy process takes time to complete. There is also a cost associated with storing the data copy, and the copy of the data may complicate security.
Now, suppose that this particular organization had implemented a hyper-converged backup platform that supports the use of copy data management. Rather than physically copying the data for the development team, an administrator could create a virtual copy of the data on the backup server. Although the developers are presented with the illusion that they are working from a dedicated copy of the data, they are actually seeing the data as it exists on the backup target. This means the virtual copy is not consuming any additional storage space.
Because the data within the backup needs to be treated as immutable, virtual data copies are implemented in such a way that any write operations are redirected to a dedicated location, thereby leaving the backup data in pristine condition.
One of the nice things about this approach to data protection is that hyper-convergence, by its very nature, is scalable. As the organization's data set grows, the backup infrastructure can grow with it.
It's still relatively new to use a backup platform as a host for bringing all of an organization's data together in one place and then making previously siloed data available on an as-needed basis. Even so, there are at least two vendors that offer products with these secondary data storage capabilities: Cohesity and Rubrik.