Why your data centre's power distribution could be your single point of failure (and how to fix it)

Data Centres
Image
October 21, 2025
4 minute read time
Image

The hidden threat

For data centre network operators and IT managers in New Zealand, a major natural disaster or a sophisticated cyberattack pose the biggest threats to your work. Of course, these events are terrible in and of themselves, but what is it about these phenomena that is so feared by this industry? It's the sudden, inexplicable outage. The shutdown without warning. The resilient servers you’ve invested heavily in, the efficient cooling, and robust UPS systems, all grinding to a halt. Oftentimes, it’s because of a critical, overlooked vulnerability: your power distribution. Even with rock-solid infrastructure and backup power, a single point of failure (SPOF) within your rack-level power can be the weak link that compromises your entire operation.

Talk to a power expert

The Problem: understanding your power distribution's vulnerabilities

A single point of failure is any component whose failure causes the entire system to fail. In a data centre environment, this can look like a basic unmonitored PDU or something as mundane as an accidental unplugging.

Here’s how SPOFs typically arise in power distribution:

The ‘blind spot’ of unmonitored PDUs
Many data centres rely on basic power distribution units (PDUs) like our own RackPower units. While our PDUs are robust and carry a strong 5-year warranty, they are unmonitored. This means they provide power reliably, but offer no data on power draw, no alerts for overloads, and no remote control, creating a significant blind spot.

Overloaded circuits and daisy chains
Without real-time monitoring, it’s easy to inadvertently overload a circuit as new equipment is added. A basic PDU offers no warning until it trips, taking down everything connected to it.

The single-cord server dilemma
While many critical servers have dual power supplies for redundancy, some equipment, particularly networking gear or older servers, only have a single power input. Connecting these to a single PDU immediately creates an SPOF.

Accidental disconnections
A bump, a loose connection, or an accidental tug on a power cord can instantly cause an outage, especially in busy data centre environments.

Lack of redundancy
Relying on a single path for power from the UPS (Uninterruptible Power Supply) to your equipment, even if it passes through a PDU, is an inherent risk.

Contact us today
Image

Five solutions to building a resilient power infrastructure with intelligence and redundancy

Eliminating these SPOFs requires a strategic approach that combines intelligent monitoring, robust redundancy, and physical security measures.

1. Embrace intelligence with RackPower’s Enlogic intelligent PDUs

Unlike basic unmonitored units, Enlogic PDUs provide critical visibility and control:

Real-time monitoring: Get granular data on power consumption at the inlet, circuit, and even individual outlet level. This allows you to identify overloads before they happen and optimise power usage effectiveness (PUE).

Threshold alarms: Set custom power thresholds. If a server starts drawing too much power or a circuit approaches its limit, you’ll receive immediate alerts via email or SNMP, giving you time to intervene before an outage occurs.

Remote management and control: Enlogic Switched PDUs allow you to remotely power cycle individual outlets. This means you can reboot a hung server from your desk, avoiding an unnecessary trip to the data centre and dramatically reducing downtime.

Environmental monitoring: Integrate sensors for temperature and humidity to ensure optimal rack conditions, preventing heat-related power issues.

Explore intelligent PDUs
Image

2. Solve the single-corded equipment problem with an Automatic Transfer Switch (ATS)

For servers and networking equipment with only one power input, an Automatic Transfer Switch (ATS) is your lifeline. An ATS connects a single-corded device to two independent power sources (e.g. two different PDUs or UPS systems). If the primary power source fails, the ATS automatically and, without a hiccup, switches the equipment to the secondary source, preventing an outage.

Shop Automatic Transfer Switch



3. Lock down your connections with Locking Outlets and Cables

Data Center Dynamics estimates that around 66% to 80% of all data centre outages are caused by human error. These accidental disconnections are easily preventable with:

Locking Outlets: RackPower’s PDUs feature locking C13/C19 outlets that physically secure power cords.
Locking Cables: Pair these outlets with matching locking power cables for an incredibly secure connection that won't vibrate loose or be accidentally dislodged.

Shop Locking Cables
Image
Image

4. Build true redundancy with A & B power feeds and dual UPS protection

Even the most advanced PDUs and secure connections can’t protect against a complete power path failure. That’s why true redundancy at every level of your power infrastructure is essential. A & B power feeds are the gold standard for mission-critical environments. This setup involves two independent power paths (A and B), each with its own PDU and UPS. Equipment with dual power supplies connects to both paths, ensuring that if one side fails due to maintenance, overload, or unexpected outage, the other continues to deliver uninterrupted power.

Here’s how to implement it effectively:

Dual PDUs per rack: Install both an A-side and B-side PDU in each rack. Connect each server’s redundant power supplies to separate PDUs.

Independent UPS systems: Ensure each PDU is backed by a separate UPS. This isolates failures and allows for maintenance without downtime.

Load balancing: Distribute equipment evenly across both power feeds to prevent overloading one side and to maximise efficiency.

ATS for single-corded devices: For equipment with only one power input, use an Automatic Transfer Switch (ATS) to connect it to both A & B feeds. If one feed fails, the ATS instantly switches to the other—no manual intervention required.

This layered redundancy approach ensures that no single failure, whether it’s a tripped breaker, a failed UPS, or a disconnected cable, can bring down your critical systems.

5. The RackPower advantage: locally tested, locally trusted

While Enlogic intelligent PDUs provide the monitoring and control, it's crucial to also rely on dependable physical infrastructure. Our RackPower unmonitored PDUs are an excellent choice for less critical applications or as part of a redundant setup:

New Zealand manufactured: Designed and built right here in New Zealand, our unmonitored PDUs meet local standards and are perfectly suited for our operational environment.

5-year warranty: We stand behind the quality of our New Zealand-made RackPower products with an industry-leading 5-year warranty, giving you long-term confidence in your infrastructure.

Locally tested: Each RackPower unit undergoes rigorous local testing, guaranteeing performance and reliability before it reaches your data centre. This in turn reduces lead times and ensures faster support if ever needed.

Start your project with us

The wrap up

Don't let overlooked weak points turn your robust data centre into an inevitable shutdown. By integrating intelligent Enlogic PDUs for elite monitoring and control, deploying ATS switches for single-corded devices, securing connections with locking cables, and leveraging the proven reliability and local support of New Zealand-made RackPower products, you can eliminate single points of failure and ensure uninterrupted operation for the foreseeable future. Ready to secure your power infrastructure? Contact us today to discuss how Enlogic and RackPower solutions can fortify your data centre.
Contact us

FAQs