Colocation Power & Redundancy: A Buyer's Guide
Power redundancy determines whether your servers stay online when a utility feed, transformer, UPS module, or generator fails. Understanding N, N+1, and 2N configurations—and how to verify them at the hall level—prevents you from paying for reliability you are not getting. This guide explains power topology, A/B feeds, generator backup, and what to demand in your colocation RFP.
Power redundancy configurations explained
N is unacceptable for any production workload. N+1 is the minimum for enterprise production. 2N is preferred for mission-critical workloads where any power interruption causes significant impact. Higher redundancy levels generally cost more because of added equipment and maintenance; get quotes to compare.
| Configuration | What it means | Reliability level |
|---|---|---|
| N | Capacity exactly matches load—no redundancy | Single failure causes downtime |
| N+1 | One backup component beyond required capacity | Survives one component failure |
| 2N | Fully redundant, two complete independent systems | Survives failure of an entire system path |
| 2(N+1) | Two independent systems, each with N+1 | Highest practical redundancy for enterprise colo |
UPS topology and A/B feeds
An Uninterruptible Power Supply (UPS) bridges the gap between utility power loss and generator startup. In N+1 and 2N configurations, UPS modules provide conditioned power and ride-through during the seconds it takes generators to start and accept load.
A/B feeds means your cabinet receives power from two independent power distribution paths (A and B). Your servers must have dual power supplies connected to separate PDUs on each feed. If one path fails, the other keeps your equipment running. Verify that the facility provides true A/B feeds from independent UPS systems—not two feeds from the same UPS bus.
Generator backup and fuel reserves
Generator backup is your last line of defense during an extended utility outage. A facility with 24 hours of fuel and a replenishment contract is different from one with 96 hours on-site. Ask for the generator test logs and fuel contract details.
- How many generators serve the facility, and what is the N, N+1, or 2N configuration?
- What is the fuel reserve on-site (typically 24–96 hours)?
- Is there a fuel replenishment contract with guaranteed delivery SLAs?
- How often are generators load-tested (typically weekly or monthly)?
- Are generators diesel, natural gas, or dual-fuel?
What to verify before signing
Do not accept "we are Tier III" as an answer to power redundancy questions. Ask for the specific N, N+1, or 2N configuration of the power infrastructure serving your hall, and read the actual SLA document for exclusions.
You can also verify power on a site visit by walking the quoted hall with the operations team.
- Power redundancy configuration (N+1 or 2N) for your specific hall—not the building marketing
- UPS topology and whether A/B feeds are from independent UPS systems
- Generator configuration and fuel reserve hours
- Utility feed diversity: does the building have feeds from independent substations?
- Power monitoring and alerting: what is measured and how are you notified?
- SLA language: what does the power uptime SLA cover, and what are the exclusions?
Power redundancy and your SLA
A facility with 2N power should offer a stronger power uptime SLA than one with N+1—but verify the SLA language. Check whether the SLA covers measurement methodology, exclusions (maintenance windows, force majeure), credit calculation, and termination rights for chronic outages. A strong infrastructure with a weak SLA still leaves you exposed.
Redundancy inside one building does not protect against a site-wide event. For that, plan a disaster recovery colocation site in a separate market.
Related Guides
Ready to compare providers?
Get matched quotes from multiple facilities or talk through your requirements with an expert.