A Reference Architecture for High-Availability Automatic Failover between PaaS Cloud Providers

As the adoption rate of Cloud Computing continues to clamber on among various application archetypes, there is a growing concern for identifying reliable automatic failover solutions between various cloud providers in an attempt to minimize the effect of recent cloud provider outages among diverse always-on and mission-critical applications in healthcare, e-Commerce and ancillary settings. Automatic failover between cloud providers stands out as a solution for course-plotting application reliability requirements in support of high-availability, disaster recovery and high-performance scenarios. Using a case study involving Microsoft's Windows Azure cloud and the Google App Engine cloud solution, we investigate some of the key characteristics in this area of concern and present a reference architecture for automatic failover between multiple Platform-as-a-Service (PaaS) cloud delivery providers in a bid to maximize the delivery of architecturally significant quality attributes pertaining to High-Availability, Performance and Disaster Recovery in a mission-critical application prototype.