Some good points. Although, I would argue a few things brought out by the author. The guy who wrote the article is Head of Enterprise Strategy at AWS so his bias leans AWS. For example, “First, rehosting takes a lot less time, particularly when automated, and typically yields a TCO savings in the neighborhood of 30%…. when you’re honest about your on-premises TCO.”…. A percentage that represents savings from hosting in AWS as opposed to ones own data center can easily be skewed. The author throws around 25%, 30%, and 52% as examples of TCO savings when organizations moved to AWS. But that percentage number is relative to a number of other things, scale being one. If we want to host 10 servers then spending $500k+ to build out 2 data centers for redundancy, etc will mean that each server will cost us a lot. If small scale, AWS makes sense. One does not need to build and manage their own infrastructure. But once an organization reaches a certain amount of required computing, storage, and networking needs then a build your own data center plan may make more sense than cloud cost wise. Also, part of the savings of a move to a cloud provider is the ability to use infrastructure as a service. (IaaS). This helps reduce administrative overhead and helps increase more efficient use of memory and CPU. If an Organization could add an IaaS layer to their data centers then that could also result in a savings. So maybe it is not as simple as saying ‘moving to AWS will save x%’ as the author says. Maybe it is more about efficiency gains attained by IaaS as compared to traditional hardware in a data center. I bet dollars for donuts similar savings could have been attained by putting an IaaS layer on top of the data center in the cases the author mentioned.