It used to be that organizations consolidated all data into one location. All a disaster recovery planner had to do, which was still no easy task, was get that data backed up and get a copy sent to an off-site location. Hybrid IT makes the challenge of disaster recovery even greater. Data, instead of being in one location is now, literally, everywhere. Organizations can have unique and critical data in secondary data centers, remote offices, Software as a Service (SaaS) applications, mobile devices, and multiple cloud providers.
The basic tenets of disaster recovery remain the same however. IT still must store a recent copy of data centrally at an alternate location other than its original one. IT needs to decide if it stores those copies centrally or distributes those copies between the various locations used by the Hybrid IT model. There are advantages to centrally storing the DR copies. It is easy to comprehend where the data is and it is easy to know where the “go to” copy of data is in case a disaster impacts a remote location.
When looking for a data protection solution that is ready for the Hybrid IT future, IT needs to ensure that the solution can scale to store all this data. It also needs to be able to scale its internal metadata databases, the data about the backups, to track the vast number of files it may be storing.
Another consideration is to ensure the proper protection of all these DR copies, as well as the unique and critical data created at the primary data center, against a disaster striking it. The options are to replicate all DR copies to another location or replicate just the primary data center’s unique data.
Replicating all the data to an alternate location means, potentially, that a large percentage of this data is protected twice, which is not necessarily bad, but could be costly. Replicating just the data unique to the production facility cuts costs by eliminating redundancy, but it does complicate recovery.
Instead of centralizing all of the DR copies into a single location, the organization may want to use a Hybrid DR model to match Hybrid IT. In this model, the data protection process creates and stores DR copies once by leveraging all the available locations. For example, remote site A may replicate to remote site B and B may replicate A. Then a SaaS application may replicate to the primary data center. The goal is to decide on and leverage the best available candidate to be the alternate site for each data set.
A Hybrid IT DR strategy utilizes resources more effectively, but it can be more complicated to manage. It is possible to simplify the complication by having a single application backup and protect all the data so that the metadata information is centralized. Then IT can go through a single interface to initiate a DR recovery regardless of the source or destination.
Hybrid IT requires that in modernizing the disaster recovery process. IT needs to make sure that it protects data across regions and locations. The organization also needs to make sure that the potential data protection solution can scale its internal metadata capabilities to meet the demands of tracking data across those multiple locations.