Two interesting failures in cloud computing occurred over the past couple of weeks. The first failure was a power outage caused by severe storms that brought down the East Coast Cloud services for Amazon (Cohen, 2012). The outage affected sites like Netflix as it disrupted streaming services for customers (Cohen, 2012). The second failure occurred when a electrical glitch brought down power in the Salesforce data center in Silicon Valley (Babcock, 2012). The outage brought down primary instances of the cloud-based service which affected customers for at least seven hours (Babcock, 2012).
Both of these cases highlight the importance of having a disaster recovery or business continuity plan. Cohen (2012) points out that in the case of AWS, not all of the fault can be placed on Amazon, as a hybrid approach would have prevented this outage from affecting customers. Just because the technology is supposed to be redundant, does not mean that having all nodes in a single location will provide redundancy. System architects need to be careful when designing cloud-based solutions to ensure that major issues affecting a particular provider do not bring down the entire application.
Babcock, C. (2012, July 12). Salesforce outage follows data center power glitch. Retrieved from http://www.informationweek.com/news/cloud-computing/software/240003577
Cohen, R. (2012, July 2). Cloud computing forecast: Cloudy with a chance of fail. Retrieved from http://www.forbes.com/sites/reuvencohen/2012/07/02/cloud-computing-forecast-cloudy-with-a-chance-of-fail/