Ask two providers for a quote on the same requirement and you will often get numbers that look very different for reasons that have nothing to do with margin. Frequently the difference is what each one counts as the space you are buying.

Data centre floor area splits into two categories, and the line between them is where a lot of commercial confusion lives.

White space is the part you occupy

White space is the technical floor: cabinets, cages and private suites, the aisles between them, and the containment that separates hot air from cold. It is the area a customer can point at. It is also the area a colocation agreement usually describes, either as a cabinet count, a cage footprint, or a figure in square metres.

The name is literal rather than metaphorical. The convention comes from the pale finishes historically used on technical floors, not from anything about how the space is allocated.

Grey space is the part that makes it work

Grey space holds no customer equipment and is where most of the capital goes. Medium-voltage switchgear and transformers. UPS modules and the battery or flywheel rooms behind them. Standby generators, their fuel storage and day tanks. Chillers, pumps, cooling towers or dry coolers. Air handling plant and the pipework that connects all of it. Fire detection and suppression. The controls and monitoring that tie it together.

A customer rarely sees any of it, which is exactly why it deserves attention during a site visit. The white space in two facilities can look identical. The grey space is where they differ.

Why the ratio tells you something

There is no single correct split, because it depends on density and on redundancy level. A low-density facility supporting a few kilowatts per rack needs far less plant per square metre than one supporting high-density AI racks. Our guide to colocation power density tiers covers how much difference that makes.

What the ratio does tell you is whether the plant was sized for the floor. A facility advertising a large technical floor with modest supporting plant has capacity it cannot energise. That is not necessarily dishonest, but it means the usable figure and the marketed figure are different numbers, and you want to know which one your contract refers to.

The question to ask is not how big the white space is. It is how much of it is currently supported at the density you need, and what has to be built before the rest is.

Redundancy does not always extend evenly

This is the detail that catches people. A facility may describe itself as 2N and mean the UPS topology. The cooling behind it may be N+1. The generators may be N+1 with a shared fuel system that is effectively N.

Redundancy is only as good as the weakest element in the chain between the utility and your rack, and that chain runs entirely through grey space. A 2N power path feeding racks cooled by a single chiller plant is a 2N electrical design in a facility that fails on a chiller.

Ask for the redundancy level of each system separately rather than accepting a single label for the building.

What this means on a site visit

Most site visits spend their time on the technical floor, which is the least informative part of the building. The floor looks much the same everywhere. Ask instead to see the plant.

Useful things to look at: whether there is physical space to add the next UPS module or chiller, or whether the plant rooms are already full. Whether maintenance can be done on one path while the other carries load, or whether a concurrent maintenance claim depends on a transfer that has never been tested. How fuel is stored and how long the site runs without a delivery. Whether the chiller plant has redundancy at the same level as the electrical path.

A facility that is comfortable walking you through its grey space is usually a facility that has built it properly.

Where the regional climate changes the arithmetic

In the Gulf, cooling plant carries a larger share of grey space than it does in temperate regions, because ambient conditions remove free cooling as an option for most of the year. That has two consequences worth knowing.

The first is that the grey space footprint is proportionally larger, so a facility designed for European ambient conditions and replicated here will be short of plant. The second is that cooling resilience matters more, because the margin between operating temperature and a thermal problem is thinner when the outside air is already hot. Free cooling and adiabatic systems in hot climates covers what is and is not available regionally.

Questions worth asking before signing

What is the supported capacity at my required density, as opposed to the total technical floor? What is the redundancy level of the electrical path, the cooling path and the generator system, stated separately? Is there physical space in the plant rooms for the next increment? How long can the site run on stored fuel? Which parts of the system can be maintained without reducing my resilience?

None of those are difficult questions for a well-built facility to answer.

The practical point

You rent white space. What you are actually buying is the grey space behind it, and the contract rarely describes that part in any detail. The ratio, the redundancy of each system and the headroom in the plant rooms tell you more about whether a facility will still suit you in three years than the floor area ever will.

Rax Data & Energy builds and operates facilities where supporting plant is specified for Gulf ambient conditions rather than adapted to them. If you are comparing colocation options and want help reading what the quotes actually cover, talk to our team.

A Real Energisation Date, Not a Brochure

Rax operates where transmission investment has run ahead of load, and we are direct about staged capacity and real dates. Talk to our team about sizing a deployment against a timeline.

Contact Rax Data & Energy

Related Articles