Thousands of companies ask themselves this question: what is better? Renew the physical servers of my company or hire servers in the cloud?
While both options offer advantages and disadvantages, lately the market does nothing but talk about the invaluable benefits of Cloud technology. But ... are these advantages really worth it? What is the cost of its acquisition?
In this article we are going to focus on the cost of acquiring a Cloud infrastructure against the same on-premise. We will analyze both options and discover what you should take into account when making a decision.
In the first place, we must know the difference between the so-called CAPEX investments and the OPEX expenses in which both types of IT infrastructures can be classified respectively.
CAPEX or Investments in capital goods:
It refers to the investment for the purchase of an asset that will provide benefits throughout its useful life, which the company will own and whose payment is not recurrent and indefinite.
In this category would be the purchase of new servers for the company: these servers are a capital asset that the company owns, which must be amortized and that often require a high investment in advance or the application for a loan.
OPEX or Operating cost:
It refers to a recurrent or continuous expense. It is usually associated with maintenance costs, consumables or certain types of services. In essence, the continuous expenses that the company must face for its operation.
In this category, Cloud Computing is generally found, which is offered on a "pay per use" basis and whose cost must be paid on a recurring basis. However, it is also the category of costs such as electricity, cooling, technical maintenance, Internet connection, etc. They must be paid if the company decides to host their own physical servers in their offices.
Referring to these two concepts we can classify, roughly, the different expenses that a company faces when opting for one or another option.
Even without being exhaustive, this table suggests the enormous amount of costs that accumulate if we want to host our servers with the same redundancy, availability and fault tolerance that Cloud servers already enjoy.
It is precisely this large number of variables that complicates a realistic cost calculation and leaves open the door to another risk that can end up being very expensive: the risk of making mistakes.
Today it is very difficult for a company to know all your IT needs for the next five years in advance. We could even say that it is impossible to predict the direction that technology will take or what future market demands will be. This rapid evolution of technological needs also increases the risk of making a wrong decision when deciding what type of infrastructure (Cloud or On-Premise) we should acquire.
The magnitude and impact of a prediction error varies drastically according to the path we have decided on.
If we talk about Cloud servers, its ability to climb is well known. That is, increase or decrease in resources according to the need of the moment. In case of mistake, it would suffice to increase or decrease the resources contracted without requiring any additional material purchase or migration.
The on-premise infrastructure, however, is limited to the physical resources that we purchased at the time of acquisition. To scale we will need to make an additional investment: acquire new and expensive servers that may involve an unexpected expense.
One way to avoid unforeseen events is to oversize our on-premise infrastructure. That is, buy in advance servers much more powerful than we expect to need. While this oversizing will prevent unforeseen events, it will also be much more expensive. As a general rule, we must consider that any underused resource corresponds to money that has been spent inefficiently.
Finally, when making a decision, we must take into account another hidden cost and that is often overlooked by a large number of organizations: the opportunity cost.
Take for example the physical space that an on-premise infrastructure will occupy: "for what other purpose could that space have been allocated?" Or, consider the time dedicated by our team to the monitoring and maintenance of the infrastructure: "in what other tasks could be invested?".
Even if we have already amortized material such as racks, routers, servers, SAIS, etc: "what benefit could we get if we sold that equipment?".
The opportunity cost measures what could have been; what we have renounced. In this sense it is important that we also note the benefits in time, money, productivity, etc. that we could obtain when opting for Cloud infrastructure.
There are reasons of special consideration for which many companies still host many of their equipment in their own facilities: office connectivity, application compatibility, etc. However, these reasons lose more and more weight as Cloud solutions become cheaper, more competitive and increase their added value.
Today we see how the profitability of a data center itself is conditioned by its scale: how much bigger, more feasible is the on-premise model. But even this reasoning is faltering when a large number of large companies have started migration to the hybrid Cloud or directly to the public Cloud.
In conclusion: the high number of costs to be considered and the uncertainty caused by a constant technological evolution mean that the acquisition and maintenance of on-premise infrastructure is a more expensive option and difficult to justify except in exceptional situations.