01Why the choice compounds
A data centre decision is not a procurement line item. It sets the ceiling on your power, density, connectivity and compliance for years, and it is expensive to undo. It pays to get it right up front.
Downtime is costly
Even a well-run Tier III facility permits planned and unplanned interruption. One outage in a poorly chosen site, with a single power feed and a weak SLA, can cost far more than the premium for a better facility ever would.
Exit is a project
Leaving a data centre means re-cabling, cross-connect migration, hardware moves and downtime risk. Choose as if you will be there for five years or more, because switching is rarely quick or cheap.
It shapes what comes next
The power headroom, rack density and connectivity you accept today constrain what you can deploy tomorrow, including AI and GPU workloads that need far more power and cooling than standard racks.
02Score every facility the same way
Shortlist objectively. Score each facility against the same weighted criteria rather than reacting to the best tour or the lowest headline price.
The weighting below is the one we start from. Tune it to the workload: a latency-sensitive trading platform weights connectivity higher, an AI build weights power and cooling higher. For candidate sites, explore real facilities on the Optronix Data Centre Map.
Interactive / Selection scorecard
Score a facility against the weighting
The bars show how much each criterion counts. Rate a shortlisted facility from 1 to 5 on each and the weighted score updates.
60of 100
Weighted score
Viable, with gaps to close in the contract or the design.
Rate the facility on each criterion to find its largest weighted gap.
A starting framework, not a scoring standard. Ratings are weighted by the bars above; tune both to your own workload and risk profile.
Sustainability is weighted “rising” on purpose. It is climbing buyers' lists as reporting obligations, and energy costs, grow.
03Tiers and redundancy
“Tier” is the most quoted and most misused word in the industry.
The Uptime Institute's classification describes a facility's capability, not a marketing badge. What matters is whether the site can be maintained, and survive a failure, without taking your equipment down.
| Tier | What it means | Redundancy | Commonly cited availability |
|---|---|---|---|
| Tier I | Basic capacity: a single path, no redundancy | N | ~99.671% |
| Tier II | Redundant capacity components | N+1 (partial) | ~99.741% |
| Tier III | Concurrently maintainable: any component can be maintained without downtime | N+1, dual paths | ~99.982% |
| Tier IV | Fault tolerant: survives any single failure without interruption | 2N / 2N+1 | ~99.995% |
The percentages are commonly cited industry references, not official Uptime Institute figures: the Tier standard defines capability (concurrently maintainable, fault tolerant), not an uptime number. For most enterprise workloads Tier III is the sweet spot, and truly mission-critical systems justify Tier IV. Be wary of “Tier III-equivalent” claims with no certification behind them.
What N, N+1 and 2N mean in practice
No margin
Exactly enough capacity to run the load. Any component failure or maintenance means downtime. Fine for development and test; not for production.
One spare
One redundant unit per system, such as an extra UPS or cooling unit. The facility can maintain or lose one component without dropping the load. The practical baseline.
Fully duplicated
Two complete, independent systems: A and B power, cooling and paths. Survives a full system failure. The standard for fault-tolerant, Tier IV-class sites.
04Power, cooling and efficiency
This is where facilities genuinely differ, and where AI workloads are rewriting the requirements. Ask hard questions about capacity, redundancy and how efficiency is measured.
Power capacity and density
Standard racks draw around 5 to 10 kW; AI and GPU racks can demand 40 to 130 kW. Confirm the per-rack limit, the total contracted power and whether there is headroom to grow, not just today's footprint.
Power redundancy
Look for diverse utility feeds (A and B), UPS with the right autonomy, generators with a credible fuel contract, and evidence of real maintenance and load testing, not just equipment on a datasheet.
Cooling
Hot and cold-aisle containment as a minimum, free cooling for efficiency and, increasingly, a clear path to liquid cooling if you will run high-density or AI hardware.
Efficiency (PUE)
A useful comparator, but ask how it is measured: annualised, at a realistic load. A lower PUE means more of your power bill goes to compute rather than overhead.
Interactive / PUE benchmark
Where a facility sits
Set a facility's annualised PUE to see where it sits against the benchmarks and how much of the power it draws reaches the IT load.
PUE
At 1.56 this facility matches the industry average, which has been flat for five years (Uptime Institute, 2024).
05Connectivity, location and compliance
A resilient facility you cannot reach, over the network or in person, or one that fails your auditors, is the wrong facility.
Connectivity
Carrier-neutral, with multiple carriers and diverse, redundant fibre paths. Check the cloud on-ramps (such as AWS Direct Connect or Azure ExpressRoute), the cross-connects available and what they cost: recurring cross-connect fees add up.
Location and jurisdiction
Latency to your users and staff, natural-hazard and flood risk, power cost and grid stability, and data residency and jurisdiction for your compliance regime.
Security and compliance
Physically: staffed around the clock, mantraps, biometric access, CCTV and zoned floors. Certifications to insist on: ISO 27001 (security), SOC 2, PCI DSS where relevant, ISO 50001 (energy) and ISO 22301 (continuity). Ask to see current certificates, not logos on a website.
SLA and support
An uptime guarantee with meaningful penalties rather than token credits, defined remote-hands availability and response times, and a clear escalation path. Read what the SLA actually commits to.
06Commercials and red flags
The headline rate hides most of the real cost. Understand the pricing model, the escalators and the exit before you sign.
Pricing model
Per kW, per rack or per square foot, and whether power is passed through at cost or marked up. Factor in cross-connect fees, remote-hands rates and set-up. Compare like for like across the shortlist.
Contract and exit
Term length, annual price escalators, notice periods and, critically, the de-install and exit terms. A good rate with a punishing exit clause is not a good deal.
Red flags worth walking away from
Vague SLA, token penalties
An uptime promise with no real financial consequence is marketing, not a guarantee.
Single feed or no generator
One utility feed, or no credible standby power and fuel plan: a single point of failure for your whole estate.
No carrier neutrality
Being locked to one network provider means captive pricing and a single point of failure for connectivity.
Opaque power billing
If they will not explain exactly how power is metered and charged, assume it is not in your favour.
No room to grow
No power or space headroom means you will be migrating again sooner than planned.
Will not show evidence
Reluctance to share current certifications or audit reports, or to allow a proper site visit, is the biggest tell of all.
07How Optronix helps
Choosing well takes a lot of technical due diligence, much of it on site, and it is easy to get wrong when you are comparing a landlord's sales deck rather than the facility itself.
Selection is the first stage of the data centre lifecycle and where we start with most clients. We help you shortlist, audit and choose data centres across the UK, Europe and worldwide, then stay with you through build and live operations.
- 01
Requirements and feasibility
We translate the workload into real numbers (power, density, space, connectivity and compliance) so you shortlist against the right criteria.
- 02
Independent facility audits
We assess shortlisted sites against the criteria in this guide (resilience, power, cooling, security) objectively, as your advisor rather than the facility's.
- 03
Selection and negotiation
We score, compare and pressure-test the commercials and SLAs, so the decision is evidence-based and defensible.
- 04
Design and delivery
Once a site is chosen, we design the white space and run the build, cabling and migration: the rest of the lifecycle with one team.
- 05
Managed services and live operations
Once you are live, we keep it running: managed services, data centre operations, smart hands and managed field services, with ongoing support wherever the site is.
We can run the comparison for you
We search 2,000+ data centres and evaluate 330+ providers on real data and live pricing, across connectivity, colocation, cloud and more, for recommendations built around your needs.
Choosing or auditing a data centre
Bring us your shortlist or your requirements.
We will run an independent assessment against the criteria in this guide. No facility has a stake in our answer.
Sources and notes
- Uptime Institute Global Data Center Survey 2024 and 2025: industry-average PUE of 1.56, flat for five consecutive years.
- Uptime Institute Tier Classification System: definitions of Tier I to IV (capability-based).
- Tier availability percentages (~99.671% to 99.995%) are commonly cited industry references (for example TechTarget and Colocation America), not official Uptime Institute figures.
- Rack power-density ranges and certification standards per general industry practice (ISO 27001, SOC 2, PCI DSS, ISO 50001, ISO 22301).
Independent guidance. This guide is vendor-neutral and recommends no specific facility or provider. Weightings are a starting framework to adapt to your own workload and risk profile, not a scoring standard. The PUE tool is simple arithmetic on the value you set.