Cydia, Foursquare, Quora … All these services have in common the same company that provides hosting some (or many) modules in the view of the page. The Amazon, which has the good fortune to bill a good money offering this technology is also blamed for a major failure that hindered the use of internet services on April 21. A week after the incident, the company went public to explain what happened.
Amazon AWS cloud computing service company, can be used for hosting entire websites or just parts of those sites. For example, the pictures of avatars on Twitter are the computers from Amazon, rather than on the computer itself microblog.
Failure according to Amazon, was human. On April 21, they initiated a change in the main service network in the United States, in order to gain scalability (ie support the demand of the moment). Therefore, they redirected some of the traffic to a network redundancy, which ended up not working very well. Result: Reddit, Foursquare, Ow.ly and several other services were affected.
In a long statement on the matter, the company says it plans to soon offer a way for customers to use different zones availability (areas within the cloud computing a site hires to use). Thus, the redundancy would be higher and would avoid this problem.
“We know how critical are our services to the enterprises of our customers, and we will do everything in our power to learn from this event and use it to bring improvements to our service,” the company wrote. As if that were not enough, she will offer discount on the bill next month Amazon EC2 for customers who have been affected. Now he spoke the language of startups.