Five today. Ninety-six on the way. That’s where Canada’s hyperscale data centres stand in 2026. As billions of dollars flow into new facilities, the spotlight is often on construction and investment. But once the ribbon is cut, success depends on operations.
These sites run 24/7, rely on complex power and cooling systems, and demand uninterrupted performance. Even a problem with the physical infrastructure supporting the IT environment can put uptime at risk, making proactive maintenance, monitoring, and operational response essential. As Canada’s hyperscale footprint grows, facilities management teams are becoming a critical factor in determining whether these investments deliver on their promise.
What makes hyperscale data centre facilities different from traditional buildings?
While most buildings are designed around people, hyperscale data centres are designed around machines.
| Traditional facility | Hyperscale data centre facility |
| Occupant comfort is the priority | Equipment performance is the priority |
| Business-hour operations | 24/7 operations |
| Planned interruptions are manageable | Downtime is unacceptable |
| Energy efficiency focus | Uptime and reliability focus |
| Limited redundancy | Multiple layers of redundancy |
| Reactive maintenance may be sufficient | Predictive maintenance is essential |
In this setting, integrated facility management for data centres doesn’t sit behind the operation. In many ways, it is the operation.
How are data centres managed by FM teams?
Data centres are managed by combining critical facilities management (power, cooling, and building systems) with IT operations, physical security, and tightly controlled procedures, all focused on continuous uptime.
From a facilities perspective, effective data centre operations rest on four areas.
1. Power
FM teams look after the full electrical chain. This includes utility feeds, switchgear, transformers, UPS systems, and backup generators. Every layer has redundancy, and every layer has to be tested regularly while the facility stays live.
2. Cooling
AI is pushing heat density to new highs. At eStruxture’s CAL-3 campus in Calgary, racks reach 125 kW and use direct-to-chip liquid cooling to keep power usage effectiveness (PUE) near 1.1. Air alone can’t cool those loads, so FM teams now need skills in liquid systems, high-pressure plumbing, and precise thermal control.
3. Critical assets
Chillers, pumps, batteries, fire suppression systems, and building automation systems all have to perform. Good data centre management depends on lifecycle planning, condition monitoring, and spotting failures before they happen.
4. Procedure
In mission-critical environments, how work gets done matters as much as the work itself. Change management, method-of-procedure documents, safety protocols, and emergency response plans keep routine maintenance from turning into an outage.
What are the challenges in hyperscale data centres facilities management?
1. Power and grid constraints
Recent forecasts suggest data centres could account for 8.6% of Ontario’s electricity demand by 2050. As AI-driven demand grows, utilities across Canada are facing increasing pressure to deliver the electricity hyperscale facilities require. FM leaders are being asked to optimize energy use, improve efficiency, and strengthen backup power strategies to maintain uptime in an increasingly constrained energy environment.
2. Cooling limitations
AI has turned heat into one of the biggest operational risks. As rack densities rise, traditional air cooling is reaching its limits, and operators are moving to liquid and direct-to-chip systems. FM teams now have to manage new technologies, new skill sets and growing pressure over water and energy use, all without compromising uptime.
3. Supply chain disruptions
When critical equipment takes months or even years to arrive, reactive maintenance is no longer an option. FM teams must plan further ahead, maintain stronger supplier relationships, and take a more proactive approach to asset lifecycle management.
4. Operational complexity and resilience
Hyperscale facilities leave almost no margin for error. Power, cooling, security and building automation must all work as one system, around the clock. As data centres become critical infrastructure for AI, cloud and essential digital services, FM is moving beyond maintenance. It’s now responsible for resilience, risk management and continuous performance.
What are the best practices for hyperscale data centre facility management?
Managing a hyperscale data centre facility now means preventing problems before they disrupt operations. Here’s what leading FM teams are doing differently.
1. Treat energy as a strategic asset
In a grid-constrained market, efficiency gives facilities a competitive advantage. Leading teams monitor power in real time and keep improving PUE. Industry-wide progress is still slow because older infrastructure and cooling limits are holding it back.
2. Shift from reactive to predictive maintenance
One in ten outages is still serious or severe, and outage costs keep rising. Predictive maintenance and sensor analytics are the AI applications operators trust most, which makes them a practical starting point.
3. Get ahead of high-density cooling
More operators are now reporting peak rack densities of 30 kW or higher. Canada’s liquid cooling market is projected to grow sixfold by 2031. FM teams that build liquid cooling expertise now will be ready when they need it.
4. Plan the lifecycle around supply chain risk
Long equipment lead times mean FM teams have to plan further ahead. Leading teams forecast when assets will reach end of life, keep critical spares on hand and build long-term supplier partnerships.
5. Standardize how work gets done
In data centre facilities with almost no margin for error, consistency protects uptime. Clear policies and standard operating procedures (SOPs) mean every shift and every technician handles maintenance, changes and emergencies the same way. They also make it easier to apply the same practices across sites.
6. Invest in people as seriously as infrastructure
More than half of operators report difficulty finding qualified candidates. Skills development helps attract and retain FM talent, and well-trained teams operate facilities more efficiently.
7. Integrate operations under one strategy
Power, cooling, security and people can’t be managed separately. An integrated FM model gives teams the shared data, accountability and visibility that critical 24/7 environments need.
Ready to future-proof your data centre operations? Explore how Dexterra’s integrated facilities management solutions can help drive reliability, resilience and performance at scale.
Frequently asked questions
A hyperscale data centre is a large facility built to support cloud computing and AI workloads, often requiring 100 megawatts of power or more. By comparison, a typical urban data centre in Canada runs at about 5 megawatts, meaning a hyperscale facility can use roughly 20 times more power.
Predictive maintenance uses equipment data and analytics to identify signs of deterioration before a failure occurs. This helps facilities teams address emerging issues, plan repairs, and reduce the risk of costly downtime.
FM teams need expertise in electrical infrastructure, high-density cooling, building automation and asset management. As cooling technologies evolve, teams also need ongoing training and clear procedures for maintenance, changes and emergency response.
Cooling protects computing equipment from overheating in data centre operations. A cooling failure can interrupt workloads and affect service availability. In such cases, facilities teams monitor thermal conditions and maintain cooling systems to support reliable data centre operations.