Disaster Recovery could be the best implementation of the ‘as-a-Service’ business model yet because it fits the cloud delivery system so well and features an entirely different value proposition than other cloud services. DR-as-a-Service (DRaaS) offers more than the added convenience of monthly payments or a less expensive way to achieve what a company is currently doing in-house. It’s an enabler, a way to get real DR functionality that most companies could otherwise never afford.
What is “as-a-Service”?
As-a-Service businesses – Backup-as-a-Service (BaaS), Storage- (StaaS), Software- (SaaS), etc – all provide an alternative to local infrastructure and local operation of an IT function. BaaS replaces backup software, a tape library, a dedupe box, etc, running in the company’s data center with data protection that’s run out of the cloud. It may include a hybrid appliance on-site, but it’s still designed to be a ‘plug and play’ implementation, one that’s paid for on a monthly basis.
Other cloud-based services also effectively move hardware and/or software from the data center to the cloud, offering a similar value proposition to the customer. By replacing these data center functions with cloud services, they convert a capital expense into an operational expense, including the implementation cost, and give the user a pay-as-you-go format. They also provide nearly unlimited scalability and the opportunity to run that particular service on infrastructure that’s typically more reliable and more secure than what they could expect to have in-house.
What is DR?
Disaster Recovery (DR), with its many definitions, has a long history in IT. Where backup stores a copy of data, DR provides the company with a method for recovering critical IT services (not just the data), in the event that a disaster (natural or man-made) brings down one of these systems – or the entire data center. The big differentiator between DR solutions is how long they take to bring the protected systems back on-line.
Recovery time equals money
Historically, this recovery time (also referred to by its planning name “Recovery Time Objective” or RTO) was directly proportional to the cost of the solution. Buying a DR system that could get you back up and running in minutes, instead of hours or days, typically involved redundant infrastructure, including software licenses. This put real DR protection out of the reach of all but the largest organizations. DRaaS is changing that.
Internal or private disaster recovery processes have been very complex and expensive to implement, starting with the need for a secondary location. But they also typically require buying more hardware and setting it up, then installing the production applications that are important enough to drive the whole DR project in the first place. You also have to keep that secondary infrastructure updated as production applications are modified, new ones are added, etc. Finally, you need to make sure that the data on the applications in the primary center is synced as rapidly as possible to the secondary location, which leads to more software and expensive WAN communications.
The “as-a-Service” model already includes the cloud, replacing a major component of the private DR process, the remote site itself. But thanks to technologies that are already commonplace in the IT world it can provide the features of an enterprise-level DR solution – at a dramatically lower cost. Server virtualization, VM image backup and asynchronous replication enable entire server instances to be captured and stored in the cloud. And recovery-in-place capabilities provides server recovery using that same cloud-based backup copy. This means that users can get disaster recovery protection, but with recovery times in minutes, for the cost that’s in the range of a typical cloud-based service.
Hybrid appliances provide an on-site copy for local recovery of backed up VMs and handle data transfer to the cloud, minimizing potential problems with latency and recovery synchronization. Plus, whatever latency that the cloud introduces impacts DR less than other cloud services, since it’s used so infrequently. For this reason, DRaaS will generate fewer objections than other cloud-based services, making it an even stronger fit.
The “as-a-Service” business model is being used for more and more IT services, from backup to application software to test and development infrastructure. But DR may be the “killer app” for the cloud-based delivery model. It uses the remote location of the cloud data center to supply the DR site and leverages server virtualization, hybrid appliances and recovery-in-place, to give smaller companies a real DR solution they can afford.