In “Hybrid Cloud Will Transform Disaster Recovery”, Greg Ness, my colleague at CloudVelox, discusses using the cloud for Disaster Recovery and its benefits as a seamless extension of the data center. This aligns well with my view on hybrid cloud requirements, which I believe are key to the hybrid cloud. Greg points out the efficiencies of cloud’s incremental cost model and how well it fits with disaster recovery, since the cloud’s infrastructure is heavily used only during local outages. One big component of cloud costs is running the cloud systems, or “instances”. In the case of DR, these instances only need to run in the event of an outage. Replicating the information from the data center systems directly into cloud storage without requiring that the cloud instances be running greatly lowers costs by ensuring the instances are not running 95%+ of the time.
Disaster Recovery is a use case where enterprises can not only reduce costs but also gain confidence in the hybrid cloud model. Hybrid cloud automation simplifies the whole DR process and allows enterprises to test it at any time.
Traditionally, the process of creating and operating a Disaster Recovery site is labor intensive, and the process of discovering, blueprinting, and replicating systems into the cloud can be very complicated and error prone. Afterwards, a group of steps called a run book needs to be generated to specify how to bring up the systems and services in the cloud when there’s an outage. Executing the run book can take hours; and that’s only if everything goes well.
With automation, the process of creating DR in the cloud becomes as simple as dragging and dropping icons and having software perform the complex steps. The group of cloud systems can be up and running in less than 5 minutes and requires only the pressing of a button. Testing now becomes easy, and enterprises can build confidence in the cloud by being able to try their application in the cloud at any time.
We’ve talked about how DR is a good use case for enterprises to reduce costs and gain confidence in this new hybrid cloud model, but the hybrid cloud operating model is more than just two clouds and a way to move applications between them. We can now extend services from the data center into the cloud allowing applications to span both rather than being bound in either. Applications requiring services that must stay in the data center can now be run in in the cloud. By enabling our local environment to extend into the seemingly infinite resources of the cloud environment we remove resource limits and enable a boundless data center. These capabilities are what make the hybrid cloud truly revolutionary for the vast majority of multi-tier legacy apps.