Recent Incidents in Data Centers Caused by Fires

In recent months, the data center sector has been marked by incidents that have drawn attention to an often underestimated risk: fire in critical infrastructure. In March 2025, a fire broke out in a large data center in São Paulo, partially affecting operations and requiring rapid contingency measures. Shortly before that, in Oregon (USA), a fire in a battery room at a data center used by social network X (formerly Twitter) also had a major impact, including global service failures.

Source: https://www.datacenterdynamics.com/en/news/fire-breaks-out-at-oregon-data-center-leased-by-elon-musks-x/

These are not isolated cases. According to a survey by broker Lockton, between 2014 and 2023, at least 22 significant incidents of fire or explosion were recorded in data centers around the world. The causes range from electrical failures and circuit overloads to problems with UPS and lithium-ion batteries.

Source: https://global.lockton.com/gb/en/news-insights/data-centre-fires-a-physical-threat-in-a-digital-world-uk

Given this scenario, an inevitable question arises: how can Tier IV data centers maintain their promise of 99.995% availability even in the face of such critical risks?

Tier IV Data Centers: Non-Negotiable Availability

According to the ANSI/TIA-942-B standard, a Tier IV data center must guarantee a maximum of 26 minutes of downtime per year (which means the equivalent of 99,995% availability). To achieve this level, it is necessary to operate with 2N electrical redundancy and total fault tolerance.

But ensuring this reliability does not depend solely on redundancy. It also requires the ability to predict problems before they develop into critical failures, which is where monitoring via an online thermography system comes in.

Online Thermography System for Data Centers: Preventing the Invisible Risk

Most fires in data centers are caused by electrical issues: overheated connections, UPS failures, battery banks, or transformers. If these critical points are not monitored in real time, small thermal deviations can turn into major disasters.

A continuous thermographic monitoring system can detect abnormal temperature rises in real time, allowing operations teams to take predictive action before downtime occurs or, in more serious cases, a fire breaks out. 

ZYGGOT^® SG: Thermal Intelligence for Tier IV Data Center Operations

Varixx`s ZYGGOT® SG was developed specifically for mission-critical environments such as Tier IV. It transforms thermal management from a corrective practice into a reliability engineering strategy, with advantages such as:

• Ultra-compact self-addressable sensors, installed directly on energized busbars.

• High-precision measurements, even under variable load conditions or high currents.

• Native integration with DCIM and BMS systems (IEC 61850, Modbus RTU/TCP, MQTT and API RESTful).

• Smart alarms and historical reports, enabling predictive maintenance and extending the life cycle of assets.

As a result, ZYGGOT® SG not only increases the MTBF (Mean Time Between Failures) of the electrical infrastructure, but also ensures compliance with standards such as NFPA 70E and ANSI/TIA-942-B.

From Prevention to Strategy

Adopting continuous thermographic monitoring with ZYGGOT^® SG brings direct benefits:

• Prevention of unscheduled shutdowns, protecting the 99,995% SLA.

• Reduced risk of fires and insulation degradation.

• Increased MTBF and greater maintenance predictability.

• Extended service life of critical equipment.

More than just meeting technical requirements, continuous monitoring represents a strategic move for data center managers seeking excellence in reliability, security, and continuity.

Conclusion

The recent fires in São Paulo and Oregon show that, even for the largest global providers, physical risk is a real threat to digital operations. For Tier IV data centers that promise near-absolute availability, investing in solutions such as ZYGGOT® SG is not only good practice, but a strategic pillar for ensuring maximum reliability in mission-critical operations.