Cloud Computing has been the talk of many as the direction for companies to move to and its image has recently taken a big hit as an error in Amazon’s data centers has caused some massive downtime for scores of websites, downing some and crippling others for at the least 11 hours. An extremely visible event, it exposes the risk that companies face when renting services from others to control own their Internet media.
Amazon.com, a company whose image largely relies on its digital abilities, suffered problems on their storage servers at around 9:41 GMT at their Northern Virginia data center during an update of the status of some online Web services. The outage affects client sites, notably like those of Quora, Reddit, Foursquare, and Hootsuite to name a few, resulting in them returning connection errors to visitors trying to access their addresses. Working with “availability zones” that are supposed to enable their customers’ websites to still function should an error occur in one of them, the trouble apparently affected many of these so-called clusters that resulted in the wide-spread disconnection.
With a guarantee of 99.95% availability for each of their data centers, East at Northern Virginia and West at Northern California, Amazon has never really fully publicly explained how their availability zones worked as an unknown number of them make up the regions, or how they are separated and would prevent simultaneous outages. The company rents out spaces on its servers to mostly start-up companies who rely on cloud computing providers to handle their data for them, exposing the risks they accept. Amazon’s Service Level Agreement, characteristically, has no guaranteed uptime for each availability zone.
Nevertheless, experts agree that the actual cloud technology is coming out more damaged than Amazon itself, giving the service around 5 to 10 years to fully recover from the blow should the interruption continue, and as companies become more wary of the uncertainty cloud service poses. This whole event is going to call into question how the relationship works between the Cloud Providers and the companies that use them.
Up until know, it has been a wall between the Cloud and the businesses using them with no access of information as to how things are functioning. With this event, the mindset is going to have to change to have a more open environment so that companies have more information as to what is going on and how things are going to be handled in the event of a down situation like this. The relationship between Cloud Providers and companies will be forced to change going forward.
Amazon has stated that they will release a detailed explanation of the problem as soon as they finish fixing it.
Charles Pullman