(To see the status of the Amazon data center, see their Service Health Dashboard)
If you spend any time online, it should not come as a surprise to you that Amazon’s Elastic Compute Cloud (EC2) data center in North Virginia has been experiencing serious technical difficulties since this morning. This has affected the availability of many major online web applications including: Hootsuite, Foursquare, Quora, Heroku, Reddit, parts of Shopify, Formspring, about.me and many more.
A list of sites affected by the disabled EC2 data center
To quote the temporary new homepage of Hootsuite, “It seems much of the Internet is struggling today due to widespread outage problems.”
Twitter, Facebook and the blogosphere have been abuzz with comments from cloud computing’s naysayers proudly announcing “I told you so” to those affected services and users; an opinion I don’t agree with.
What makes this outage significant is the size and number of services that have been affected. If these companies had chosen to host their services with another web/database/server hosting company, and that company ran into issues, the end result would be the same.
In fact, I would argue that smaller hosting companies are more susceptible to this than large cloud service providers like Amazon (EC2), Google (AppEngine), and Microsoft (Azure). They are more likely to have less backbone redundancy, less geographic distribution of their infrastructure and have less resources to (money and expertise) to assign to dealing with an outage.
I think the lesson here is for services that cannot afford downtime to mirror their web applications in more than one data center, located in different geographic regions. Services like Microsoft’s SQL Azure Data Sync, for example, will help make this an easier process.
As of the time of posting this entry, the EC2 servers are still having problems. This is a bad day for Amazon, and for those who reply on its North Virginia data centre. I just hope it doesn’t scare away potential users of cloud computing.