It seems like every week or so there’s a report of a major data center outage. Major events like Hurricane Sandy remain an important lesson. Even things like the Amazon Web Services outage earlier this year remind us that merely trusting in the biggest providers is not enough to assure uptime in the event of an outage.
It’s been clear since colocation first became an option that geo-redundancy (GR) was the best way to assure that when disaster strikes, your data stays safe, and stays available.
The main issue with geo-redundancy is that it adds complexity and cost to your architecture. Because of the obstacles associated with a truly reliable GR failover system, many companies pursue more economical approach that may not operate as seamlessly, but still provides an emergency fallback.
The Drawbacks of Geographic Redundancy
GR sounds simple because it is simple: Don’t put all your eggs in one basket.
The trouble with that is that most companies are not in control of all the eggs they rely upon on a daily basis. For instance, as more and more companies begin to rely on third party applications, it might be that an outage in Omaha takes out the billing software that you use. Or that a tornado in Texas means that your appointment booking software doesn’t work.
Like other types of security, you can only be as prepared as the least reliable aspect of your ecosystem.
So if the question is, “how important is GR?” the answer needs to include the viewpoints of all the vendors you rely on as well. If they don’t have GR, and you do — well, the truth is, you don’t. It’s too easy to think you are covered when you are not.
Over-Reliance Is a Mistake
So geo-redundancy is great, but it’s only one part of the puzzle of reliable uptime. You need to make sure that you have your critical systems highly available in multiple locations. This may involve more convoluted provisioning, which suggests the need for a better toolset, perhaps something like DCIM software to provide more oversight and control, or a third party vendor tasked with that oversight.
Geo-redundancy is one thing, but it won’t matter if your provider has mismanaged infrastructure and switching. Most people have no direct control over their network and rely on third parties to keep the facility operating and connectivity up, so it’s critical that these parties have tested your backups and failovers before it’s necessary.
Geo-redundancy is more important than ever. To protect our customers, Atlantic Metro maintains network PoPs and colocation facilities across the country. Furthermore, we offer cloud-based Business Continuity solutions with regular failover testing to keep your data safe, regardless of where disasters may strike. If you’re interested, you can contact a sales representative here.