Data Centers: What they are (and what they are not) – Part I

A short educational series focused on the physical infrastructure of Data Centers, designed to clarify concepts and separate facts from interpretations.

I have noticed a growing number of LinkedIn posts (and content on other social networks) about Data Centers. With the recent and announced investments in Portugal and Spain, it is understandable that the topic is receiving increased attention.

I have worked in Information and Communications Technologies for almost 40 years and, throughout much of that journey, I have been connected (directly and indirectly) with organizations that operated — and still operate — Data Centers. I was also a university lecturer and taught ICT-related subjects for 11 years as an Invited Assistant Professor.

There is excellent content available on this topic, but there are also many questions and differing interpretations. For that reason, I have decided to publish a short series of educational articles aimed primarily at non-technical readers. I see this initiative as a modest contribution to helping shed some light on the subject.

I will try to answer questions such as:

• What is a data center, really?
• What is it used for and who uses it?
• What objective benefits does it provide?
• Who can use it and how does pricing work?
• What impacts and externalities should we consider?

Let me start by addressing a common misconception: the idea that a Data Center is primarily an investment in data-processing technology.

In practice, a Data Center is a real estate and infrastructure investment: a facility (purpose-built or adapted) designed to house computers. Much like a hotel building is an infrastructure designed to accommodate guests.

This distinction is important. Just as in the hospitality industry, where the property owner, service provider and guests are often different entities, in the Data Center world the following roles may also be performed by different organizations:

• The entity that builds, owns and maintains the physical infrastructure (the building itself);
• The entity that owns, operates and maintains the IT systems and provides computing services;
• The entity that consumes the computing capacity.

A good example frequently mentioned recently is what many refer to as the “Microsoft Data Centre in Sines”. In practice, the roles are separated:

• Building / Physical Infrastructure: Owned and operated by Start Campus.
• Computing Infrastructure: Owned, operated and maintained by Nscale.
• Consumption of Computing Capacity: Microsoft and its customers (through a multi-year agreement between Microsoft and Nscale).

This is why, when people refer to a “Microsoft”, “Amazon” or “Google” Data Center, they are not always referring to a building owned by those companies. In most cases, the Data Center owner is an independent organization, and multiple companies may operate computing environments within the same facility.

Throughout this series, whenever we refer to a “Data Center”, we are referring to the physical infrastructure where computers are installed or, in information systems terminology, hosted.

That is all for now. See you in the next article.

By Paulo Simões, Founder of Cloud365