Audits and planning
By testing the built site against what each Tier requires, path by path, rather than counting equipment. Uptime Institute defines four levels, from Tier I (Basic Capacity) to Tier IV (Fault Tolerant), and is explicit that adding components does not by itself reach any Tier, because distribution paths and other elements are evaluated too. An audit finds where a site claimed as Tier III or IV would not meet the definition, without certifying it.
Because a claim is easy and a design review is not. Uptime Institute notes that it is easy to claim Tier compliance, and a different matter to lay a solution open to a rigorous review. It is also the only organization permitted to certify data centres against its Tier Classification System, so anything else, including this kind of audit, is a check against the published definitions rather than a certification. The check matters to owners buying colocation, to tenants planning maintenance windows, and to operators whose design has changed since it was built. The definitions themselves are explained in data-centre Tiers; this page covers how a built site is tested against them.
A stepwise build-up, from a UPS and generator to fault tolerance. Uptime Institute's own summary: Tier I includes a dedicated IT space, a UPS to filter spikes, sags and momentary outages, dedicated cooling, and an engine generator. Tier II adds redundant critical power and cooling components. Tier III adds a redundant delivery path, so that no shutdowns are needed for equipment replacement and maintenance. Tier IV adds fault tolerance, so that equipment failures and path interruptions stop short of the IT operations.
| Tier | What it adds | What the audit tests |
|---|---|---|
| Tier I | UPS, engine generator, dedicated cooling | That each exists and serves the critical load |
| Tier II | Redundant power and cooling components | That each component can be taken out with spare capacity left |
| Tier III | A redundant delivery path | That every component and path can be shut down for work without affecting IT |
| Tier IV | Fault tolerance | That a single failure anywhere stops short of the IT load |
Because the paths matter as much as the boxes. Uptime Institute calls Tier Certification a performance-based evaluation, not a checklist, and states that increasing the component count does not determine or guarantee any Tier level, because distribution pathways and other elements are evaluated too. Its own example is Tier II: it allows concurrent maintenance of capacity components such as engine generators, UPS, chillers and pumps, but not of switchboards, panels, transfer switches, transformers, bus bars, cables and pipes. A site with redundant UPS modules but one switchboard downstream may still need a computer-room shutdown to service that switchboard. That is the most common gap between a claim and the build.
Each item ties back to the UPS. A UPS pair is only as redundant as the paths below it, and an audit that stops at the UPS output misses most of what Tier III and Tier IV require.
With transfer switches, and with checking. Uptime Institute's Tier Standard includes a concession for equipment with odd numbers of cords (1,3,5) in the form of rack-mounted transfer switches, but Tier III and Tier IV sites must still have multiple, independent feeds to the rack. Miscording is the quieter problem. Uptime Institute cites anecdotal evidence that 1-10% of servers in a data centre may be improperly corded, with both cords plugged into the A distribution, so the server goes down with side A despite having two power supplies. A sample check during the audit finds them; rack PDUs with outlet-level monitoring make them easier to spot afterwards.
Because it is a single point that can drop everything. Uptime Institute reports that accidental EPO activation is a recurring cause of downtime, and its Tier Standard requires that the EPO system can be maintained, isolated or removed without affecting the critical load in Tier III, with Tier IV also requiring a fault-tolerant EPO system. An audit checks how the EPO is wired, what it shuts down, how it is protected against accidental activation and how it is tested.
No, and neither do the Tiers. Uptime Institute removed references to expected downtime per year from the Tier Standard in 2009 and notes they were never part of the definitions. What an audit can say is which single failures or maintenance tasks would affect the IT load today, and what it would take to remove each one. That list is more useful to an owner than a percentage.
Yes. Send the claimed Tier or redundancy level, the single-line diagram and the UPS and generator list through the audit quote request. The power path is covered directly; cooling and fire-protection findings go into the same report and are routed to the matching specialist service. For the UPS side of a data centre, see UPS maintenance and repair for data centres.
No. Uptime Institute is the only organization permitted to certify against its Tier Classification System; an audit checks the site against the published definitions and lists the gaps.
No. Uptime Institute states that component count does not determine any Tier; Tier III also needs a redundant delivery path so every component and path can be maintained without affecting IT.
Tier II allows concurrent maintenance of capacity components only; Tier III adds a redundant delivery path, so switchboards, panels and cables can also be maintained without a shutdown.
Uptime Institute cites anecdotal evidence that 1-10% of servers may have both cords on the A distribution.
Sources, checked 2026-10-04:
Tell us the site, the UPS and the timing. We reply and quote within 4 business hours, or call 1 (438) 800-0382
Brand names are trademarks of their owners and are used only to identify the equipment we service. We are not the manufacturer and are not affiliated with it.