No organization looks forward to the day they implement their disaster recovery (DR) plan. But like any good insurance policy, DR is an essential component of business continuity and preservation.
Most people understand that much. Where organizations often fall short, however, is the details and preparation. The only thing worse than having no plan at all is not going through the proper research and review prior to the event that takes you offline.
In other words, don’t be like Kip and Gary in the above comic.* Start the process now, and keep in mind these six, baseline items:
1 – Identify and plan for your most critical assets. Because of the high resource requirements of a good DR plan, focus only on the processes and applications that are most crucial to your business while you restore normal operations. For many companies, those may be customer-facing applications and systems like e-commerce sites or portals. Applications like email, while important, may take a secondary position, and systems for internal use only – like HR or accounting applications – may fill out a third tier.
2 – Determine RPO/RTO. The Recovery Point Objective (RPO) is the maximum amount of time your business can tolerate between data backups. If your RPO is one day, that means you can survive losing one day’s worth of data, but no more. Your Recovery Time Objective (RTO), on the other hand, is the target for restoring regular service after the disaster strikes. Neither metric is arbitrary, and you’ll likely have to crunch a lot of numbers and coordinate with virtually every business unit to determine the most accurate objectives.
3 – Scope out the technical mechanics. The hybrid era of IT, for all its benefits, only makes DR planning more difficult. Critical business processes and applications exist in a complex web of interdependencies. You’ll have to map relationships across server, storage and network infrastructure and develop accurate scripts to ensure apps function like they’re supposed to in the recovery environment.
4 – Select an appropriate failover site. Traditional DR requires redundant infrastructure in which to failover. Not only is this pricey, you have to choose a site that makes sense geographically (i.e. – low odds of being affected by same event) and offers an SLA that’s up to your current standards.
5 – Take advantage of the cloud. For many organizations, designing a robust DR plan is significantly impeded due to extremely high cost and resource requirements. Disaster Recovery as a Service (DRaaS), however, is a cloud-based solution that eliminates the heavy capital expense, putting DR within reach of companies unable to acquire the redundant infrastructure needed to restore service. INAP offers a comprehensive DRaaS solution for hosted private cloud customers that includes seamless failover and failback, an RPO within seconds, and an RTO within minutes.
6 – Document, Test, Refine. This is the hat trick of ensuring effective execution. Each of these components is critical. Your plan needs to be specific and detailed. Plans, procedures, responsibilities and check lists should be clearly documented. You want your team to have clear marching orders and leave little to interpretation in the middle of a crisis.