Forces of Nature and the Power of Planning
You’ve probably heard that this past weekend’s severe storms ravaged the DC metro area and caused devastation to many in the region. What you may not be aware of is that the disruptions had repercussions well beyond that. Since DC is THE HUB of global Internet traffic, all aspects of the online economy were affected. The storms interrupted the service of many popular sites such as Instagram, Pinterest, Netflix, Heroku to name a few – and many were still struggling to recover hours later after they went down. The outages, understandably, resurfaced the always lingering concerns about Cloud reliability and disaster recovery, among other things. Media pundits immediately released stats about the various cloud providers’ outages in the past year and why this is bad for you as a consumer - you know, pretty much the same repeat wave of media frenzy that we see every time the Cloud (pardon the pun) is down.
You all know the most publicized cloud benefits: faster time to market, allowing the rapid scaling of resources, the enablement of both business executives and IT staff allowing them to refocus on the specific strategic aspects of their roles, etc. . But despite all the proven benefits, perceived data vulnerability is still the top concern, and it continues to be a major impediment to a wider and faster cloud adoption. And Mother Nature does not make it easier at times.
Which brings me to my point – Cloud can be, and will still be, the better option!
Why? If a server goes down in your own data center can you hear it “fall”? Do you have the 24x7 support staff to stand up your system, or the sufficient power sources to keep your machines up and running (and air-conditioned) when entire metropolitan areas are without power for days? Do you have the kind of data redundancy needed for mission critical operation (including managing claims for millions of customers)? What are your data redundancy, business continuity and disaster recovery plans? If you haven’t thought about these concerns, now is probably a good time to start planning.
As you ponder these crucial issues, keep in mind that all clouds are not created equal. Do your research on your cloud vendor and ask them questions around these items:
- Make sure that your cloud provider employs a disaster recovery (DR) architecture that ensures that the disaster recovery point is less than 15 minutes, and disaster recovery time is less than an hour, or your business will suffer.
- Ask what their standard business continuity practices are, and how they are supplemented with dedicated preparation for catastrophic events.
- Find out what their current response plans are for a series of disaster scenarios, and if they test their response in production by simulating disasters.
- Inquire if your cloud provider offers flexibility to place instances within multiple geographic regions and zones, each with its fault separation. This means that zones are physically separated within a typical metropolitan region, on different flood plains, and/or in seismically stable areas. In addition to discrete uninterruptable power source (UPS) and onsite backup generation facilities, they each should feed via different grids from independent utilities to further reduce single points of failure.
Data centers are designed to anticipate and tolerate failure while maintaining service levels. Does your business have the scale to match that?
Find out, before it’s too late.