Cloud  computing has been the “next big thing” for the past couple of years, with every tech company under the sun trying to figure out a way to leverage cloud technology and every service provider looking for ways to support it. Cloud technology is pretty cool and can really save the day, when it works.

But as with all technology, the question isn’t if it will fail, but when. Many people have spent the last two years re-thinking their infrastructure, moving critical servers and server backups to the cloud, where they will be “safe.” That’s a recipe for disaster, because, as CRN has highlighted very succinctly, there have been many headline-grabbing cloud outages in the first half of this year alone. If you bought into the hype completely and you’ve put all of your eggs in the cloud basket, you may find yourself in serious trouble with a lot of broken eggs.

The remedy for this should be obvious. Rather than move all your critical server operations to the cloud, you should continue to keep a near-line server backup onsite, as your first line of defense. Companies that have a local backup – in addition to their cloud backup – don’t lose any sleep over cloud failures. It’s a blip on the screen and barely worth noting.

Let’s consider a scenario that’s a compound failure: your local server goes down and your cloud provider lets you know you can’t failover into the cloud because their data center was hit by lightning last night, which is exactly what caused a recent Amazon cloud failure in Europe. This is precisely why you need near-line backup as well. IT administrators know if they’re using a best-in-class business continuity solution like ShadowProtect, they can get servers running again in a few short minutes. If they’re also running StorageCraft’s HeadStart Restore technology in conjunction with ShadowProtect, users would barely notice the server went down.

So when putting together or revising your disaster recovery plan, save yourself some headaches and only give one ear to the hype. Make sure your plan covers multiple failures and includes near-line and off-site backup.