Steven Brown, Vice

President, EcoStruxure IT Line of Business, Schneider Electric, on why connecting power, cooling, buildings and IT through Energy Intelligence will be critical to the operation of AI-native data centres. rtificial intelligence is changing

For decades, operators could rely on relatively stable environments. Power demand was predictable and cooling systems were engineered around known workloads. Operational technologies such as building management systems( BMS), electrical power management systems( EPMS) and data centre infrastructure management( DCIM) software each performed their role effectively, largely within their own domain. However, that operating model is rapidly reaching its limits.

Power swings

Today’ s AI clusters introduce highly dynamic workloads, unprecedented rack densities and emerging power architectures that require infrastructure to respond continuously rather than periodically. At the same time, operators face growing pressure to improve resilience, optimise energy use and make faster operational decisions, often while managing increasingly complex facilities with fewer specialist resources.

The challenge is no longer a lack of data. Modern data centres already generate vast amounts of operational information. The real challenge is turning that data into intelligence that understands how the entire facility operates as one connected system.

This requires a different approach – one that brings energy technology and intelligence together in a single platform.

From isolated systems to connected operations

Historically, separating operational technologies into silos made perfect sense.

BMS controlled environmental conditions, EPMS monitored power infrastructure, DCIM

24

INTELLIGENT CIO EUROPE www. intelligentcio. com