In mid-April, Amazon EC2 services had a small problem. Apparently, a large swath of its cloud computing environment had such substantial trouble that a number of customers had server issues. A number of high-profile sites, including Reddit, Evite, and Foursquare, went down when Amazon experienced issues in their US East 1a region (Justinb 2011). The author's library was heavily invested in the affected availability zone, and although they were unable to launch new instances, they managed to get through without downtime. To design a sufficiently redundant system in a cloud environment, one should focus on three primary areas: (1) system configuration; (2) data replication; and (3) service fail-over. This article briefly examines each of these areas.