Intuit’s cloud services went down last week for an astonishing 36 hours. That’s a long time if you’re a business running your books off of the cloud using Intuit products. According to a statement by Intuit, the outage occurred during routine maintenance when a power failure brought down their primary and back-up servers. This raises some legitimate questions about how Intuit is implementing its cloud solution, but it also was a high profile example of the dangers inherent in cloud computing. The real question though in my view is: Could this happen in-house? And the answer is: Of course it could. If it could happen to a cloud provider (which is after all just a data center just like any that you find inside an enterprise), then it could happen anywhere.
Let the Critics Come Forth
In a letter on the Intuit web site from Intuit president and CEO Brad Smith, he apologized for the outage. Comments on the page suggested an under-current of anger and frustration from some customers. One legitimately asked why the primary and backup systems were housed in the same place. That’s a very good question and one cloud customers may want to ask cloud providers in the future when signing service-level agreements. There seem to be some open questions about how Intuit got into this mess and how they could have prevented it.
Those are valid and reasonable questions to be asking after an outage like this, and the individual company should be held accountable for giving customers answers, but my issue is when people use a single outage such as this to open the FUD (fear, uncertainty and doubt) flood gates to use an outage to condemn cloud computing as a notion because of an isolated incident. One commenter on the Ziff-Davis site wrote:
“This outage just illustrates my point that so-called cloud computing is inherently unreliable. Unfortunately the computer industry is incredibly trendy and once some experts declare the cloud is the way to go, the rest follow without thinking.”
In my view it does no such thing, anymore than when a system goes down in-house, that it demonstrates the unreliability of enterprise computing.
Outages Are Going to Happen, Folks
Outages are going to happen. That this site was down for 36 agonizing hours had to be awful for individuals that depend on this data to run their businesses. I’m not trying to sugar coat that or make excuses for Intuit. What I do want to do is point out that outages will happen in any data center anywhere for any number of reasons, some of which are in the control of the center managers and some which are not.
You can and should learn from these incidents and I hope Intuit uses this as an opportunity to revisit their set-up and policies to prevent a similar incident from happening again. If they don’t, they will risk losing their customers moving forward, but let’s not pretend that this one incident reflects on cloud computing model as a whole because if it does, tell me that the next time your Exchange server goes down for a half a day. It happens and it will continue to happen. But let’s be intellectually honest about this, and not try to use it to put forth an anti-cloud agenda because it really isn’t a valid argument.