Cloud cost view

The advent of cloud has all but eliminated the limitations provided by each of the preceding points. Cloud providers now completely manage the undifferentiated heavy lifting for their customers. They've bought the physical space to house the data centers; deployed network, electric, and cooling systems to support millions of machines; managed security of the physical infrastructure, performed background checks, and segregated duties to ensure that no staff has both physical and logical access to systems; designed and deployed state-of-the-art cooling systems; and have a staff of focused operations engineers to maintain their massive fleets of data centers. For a further discussion on the cloud providers' scale, please refer to Chapter 5, Scalable and Available.

In every one of these points, the CSPs aggregate and reduce costs through economies of scale. A company such as AWS or Azure can spend more time designing more secure systems than other enterprises because they have more resources, money, and commitment for ensuring security. In the case of cooling, CSPs can apply greater engineering know-how to developing greener, more efficient cooling systems than any other enterprise on the market. For example, Google is using seawater to cool its data centers in Finland, which reduces stress on local freshwater sources and local electric generation, and increases efficiency. You can refer to https://www.wired.com/2012/01/google-finland/ to read more about this.

 Apart from the tangible benefits of adopting cloud, there are several intangible cost benefits (opportunity costs), which are enumerated as follows:

These tangible and intangible cost benefits are enormously beneficial to every single organization with an IT department. This is explained well by the upswing in cloud adoption that started in 2013 by large enterprises. The collective realization dawned on these entities that the cloud has many upsides and almost no downsides to adoption—the sole valid exception being the unrealized return on capital investments made recently in the IT infrastructure (for example, Company X just invested $50 million in a new data center that came online last month). Unless Company X can find a way to relinquish their new data center without recuperating commensurate sunk costs, the math is likely not in favor of cloud adoption until the next refresh cycle (this term is used to describe the time in-between the replacement of computers and associated computer hardware).