5 Disaster Recovery Misconceptions

Ben Miller (Profile)
Wednesday, June 5th 2013

Do you know how your business technology would fare if a true disaster were to hit? With the rate technology and your applications change and evolve, your DR plan may need a dusting off and updating. If your plan is outdated or relies on older assumptions, you may have gaps in your protection.

Don’t leave your infrastructure vulnerable. Assess your plan for the most common misconceptions of disaster recovery.

Misconception #1: Backup-as-a-Service and Recovery-as-a-Service are the same.

A good DR plan is not about backups, but rather it’s about getting back up and running as quickly and efficiently as possible. The placement of that one space makes a big difference.

Backups and Backups-as-a-Service traditionally are concerned about the data within an application. The recoverable pieces are much like a jigsaw puzzle, they need to be put back together in a usable way, otherwise they’re just data. If you’re a business unit owner, imagine your excel application crashes. You have the file backed up, but you have no way to access or run that file because you need the application to do so. Backups will be a component of a complete disaster recovery solution, but they’re not the complete answer.

Recovery-as-a-Service is about protecting the entire application and everything that’s inside of it. Cloud-based Recovery-as-a-Service should use protection groups to ensure the safe recovery and protection of every single virtual machine (VM), virtual application (vApp) and piece of data within your application. Each is recovered at the same point in time, ensuring a quick stand-up of your application post-declaration. There is no putting together of the puzzle, the puzzle is saved as one complete picture.

Misconception #2: One size Disaster Recovery plan fits all applications.

One type of DR solution is not necessarily the right fit for every application you have. Some applications that don’t have a high rate of change or that aren’t business critical could possibly withstand only backups as their recovery plan. If you can afford a few days to a few weeks of downtime for an app, this is an excellent place to trim the budget and use backups.

Others may require an RPO and RTO much shorter, in the window of minutes or less, for high rate of change applications that are business or mission-critical. These will cost more, but there are ranges of affordability linked to the RPO and RTOs that each application requires.

The key here is that as long as money is finite for your organization or business unit, each application needs to be evaluated on its own for the best allowable RPO and RTO. Those rankings and answers will correlate to price, which should be stratified in comparison to other applications. A truly economic and well-thought-out DR plan will have varying levels of protection, from cloud-based DR to backups for each application. Because if you’re protecting every application the same way, you’re likely either paying too much or not protecting your critical applications enough.

Misconception #3: Most disasters are natural.

You may be planning for the tornado, or the hurricane, or even the earthquake; but that’s not what has the highest likelihood of impacting your datacenter. Most disasters are not natural, they’re technological, human or random error.

It can seem easier to plan for the meteorological disaster than planning for when someone pushes code that is three versions old, but the reality is that more declarations occur from unnatural disasters than those that occur with the benefit of the weather channel’s advance warning.