There have been a number of high profile “outages” in the Cloud recently.  Google was offline for almost two hours last week for about 14% of its users (millions of individuals),  GoGrid was nearly shut down a month ago due to a widespread DDOS attack on its services and Amazon’s EC2 has had intermittent problems with connecting to its virtual, rented computers in its Elastic Computing Cloud.

Wasn’t the Cloud was supposed to be scalable?  Wasn’t Cloud supposed to be resilient? Why is it having so much trouble and why would I ever trust my data in the Cloud?

Here’s my thinking:  Cloud >= stability of any IT network

Proof?  I could do research on average network uptimes in small and medium businesses (which I’ll define as network that supports fewer than 10,000 people). I bet if I did that research that the best networks in that category have only 99.0% uptime.  That’s more than 7 hours of down time every month.  Sure, most of the downtime in this category is probably planned maintenance, but it is downtime nonetheless.  Amazon’s EC2 uptime is 99.96%, which is about 10 minutes of downtime per month.  That’s a big difference in the favor of Amazon’s Cloud.  I am confident that if we looked at all Cloud providers in the aggregate, we would come to similar conclusions.

For the hassle, I would much rather outsource the maintenance of my network infrastructure to experts (who have a bench deeper than my entire staff) who have every motivation to keep uptimes as high as possible.  

Is the Cloud perfect?  No, but it is just as good as our current system and liable to only get better.