BMS for Data Centres and Server Rooms: Controls for Critical Infrastructure

A data centre BMS maintains tight thermal envelopes for IT equipment, sequences CRAC and CRAH units for efficiency, manages free cooling, integrates leak detection, and generates Power Usage Effectiveness data. It must hold inlet air within ASHRAE A1 or A2 limits and alarm before thresholds are breached, not after.

Why does a data centre need a specialist BMS, not a standard commercial one?

A data centre is unlike any other building type. IT equipment runs continuously, generating concentrated heat loads of 5–25 kW per rack. Server hardware is sensitive: an inlet temperature excursion above 35°C can cause thermal throttling, hardware faults, or unplanned shutdown. There is no tolerance for the kind of temperature drift that might be acceptable in an office or warehouse. At the same time, data centres are significant energy consumers — cooling can account for 30–40% of total facility power — and operators are under increasing pressure to reduce Power Usage Effectiveness (PUE).

The BMS is the control layer that makes this possible: maintaining tight thermal envelopes, sequencing CRAC units for efficiency, managing free cooling opportunities, integrating leak detection, and generating the data needed for PUE reporting. Alpha Controls delivers BMS solutions for data centres and server rooms across London and the South East. For a deeper look at why a standard commercial BMS configuration is the wrong tool for a data centre environment — and what a purpose-configured system looks like — see our article on BMS for data centres: why standard building automation isn't enough.

What are the ASHRAE A1 and A2 thermal envelopes?

ASHRAE TC 9.9 defines the A2 thermal envelope applicable to most enterprise-class IT equipment: an allowable inlet temperature range of 10°C to 35°C with a maximum rate of change of 5°C per hour — both parameters that require continuous monitoring, not periodic spot checks. Most enterprise equipment is rated to ASHRAE A1 or A2 class:

• ASHRAE A1: inlet air temperature 15–32°C, humidity 20–80% RH non-condensing (recommended: 18–27°C).
• ASHRAE A2: inlet air temperature 10–35°C, humidity 20–80% RH non-condensing.

Breaching these envelopes has real consequences. Temperatures above the upper limit cause thermal throttling — processors reduce clock speed to protect themselves, reducing compute performance. Sustained excursions cause hardware failure and void manufacturer warranties. Low temperatures are less immediately damaging but increase the risk of condensation if humidity is not also controlled. The BMS must alarm before these thresholds are approached, not after they are breached. EN 50600 — the European standard for data centre facilities and infrastructure — defines environmental monitoring requirements under its availability classification framework, including mandatory temperature and humidity monitoring at equipment inlet level, not just at room level.

How does a BMS integrate with CRAC and CRAH units?

Computer Room Air Conditioning (CRAC) units and Computer Room Air Handlers (CRAH) are the primary cooling plant in most data centres. Modern units expose their control interfaces via Modbus RTU or BACnet MS/TP, allowing the BMS to read and write key parameters:

• Read points: supply air temperature, return air temperature, cooling coil leaving water temperature, compressor status, fan speed, unit fault status, filter pressure drop.
• Write points: supply air temperature setpoint, fan speed override, unit enable/disable.

With this integration, the BMS can coordinate multiple CRAC units as a system rather than each unit acting independently. This eliminates the "fighting" that occurs when independently controlled units simultaneously heat and cool different zones of the same space — a common source of both hot spots and energy waste in server rooms without BMS integration.

How does the BMS manage hot aisle/cold aisle containment?

Hot aisle/cold aisle containment is now standard practice in well-designed data halls. Server racks are arranged so that equipment exhausts hot air into contained hot aisles, while CRAC units supply cold air into contained cold aisles. The BMS role in contained environments is to maintain the correct pressure differential across the containment barrier:

• Cold aisle containment: the cold aisle is sealed above. The BMS maintains slightly positive pressure in the cold aisle to ensure cold air enters server inlets rather than recirculating.
• Hot aisle containment: the hot aisle is sealed. The BMS ensures adequate extraction from the hot aisle to prevent hot air migration into the cold supply path.

The BMS sequences CRAC fan speeds using variable frequency drives to maintain these differentials, adjusting dynamically as IT load (and therefore heat output) changes throughout the day.

What is PUE and how does the BMS measure it?

Power Usage Effectiveness is the primary energy efficiency metric for data centres: PUE = Total Facility Power / IT Equipment Power. A PUE of 1.0 is theoretical perfection (all power goes to IT). Legacy facilities often run at 2.0 or above. A modern well-managed facility should target below 1.4; hyperscale facilities achieve 1.1–1.2.

Achieving and demonstrating target PUE requires a sub-metering strategy integrated into the BMS. For a detailed guide to how energy meters connect to a BMS and what granular consumption data enables, see our article on energy metering and sub-metering.

The sub-metering strategy for PUE monitoring covers:

• IT load metering: power meters on PDUs (Power Distribution Units) feeding server racks — measuring actual IT load in real time.
• Cooling load metering: power meters on each CRAC unit, chiller, cooling tower, and pump.
• Lighting and ancillary loads: measured separately to understand the full facility load breakdown.

The BMS calculates live PUE from these inputs and trends it over time. This data drives operational decisions: when to enable free cooling, how many CRAC units to run, and whether capital investment in more efficient plant is justified.

How does BMS leak detection protect a data centre?

Water in a data centre is a catastrophic risk. Cooling systems use chilled water, condenser water, and sometimes direct expansion refrigerants — all under pressure in the same space as sensitive IT equipment. A leak under a raised floor may not be visible until significant damage has occurred.

BMS-integrated leak detection addresses this through:

• Rope-style sensing cables routed under raised floors along water pipe routes — detecting the presence of water at any point along their length.
• Point sensors in drain sumps, under precision air conditioning units, and at pipe penetrations through walls.
• Automatic shut-off valve integration: on detection, the BMS can close motorised isolation valves on chilled water circuits serving the affected zone, limiting the volume of water at risk.
• Immediate alarm escalation: leak alarms are Priority 1 — they page on-call engineers immediately regardless of time of day.

Alpha Controls provides leak detection installation as a standalone service and as part of integrated data centre BMS projects.

When can a data centre use free cooling?

Mechanical refrigeration is the largest single energy cost in most data centres. When the outside air temperature is low enough — typically below 12–15°C wet bulb for UK climate — the BMS can switch from mechanical cooling to free cooling (economiser mode), circulating cooled water or air without running compressors.

The BMS manages this transition automatically:

• Continuously monitors outside air wet bulb temperature (or dry bulb as a proxy for air-side economisation).
• When free cooling conditions are met, opens economiser valves and modulates cooling tower or dry cooler fans to achieve target supply water temperature.
• Keeps mechanical refrigeration on standby, ready to start if outside conditions change or IT load rises beyond free cooling capacity.
• Manages the transition smoothly to avoid supply temperature transients that could cause IT equipment thermal stress.

In the UK climate, free cooling is available for a significant portion of the year. A properly configured free cooling strategy can reduce annual cooling energy consumption by 30–50%.

What is N+1 CRAC redundancy and how does the BMS manage it?

Critical data centres are designed with N+1 cooling redundancy — one more CRAC unit than the minimum required to handle the IT load. The BMS must manage this redundancy actively:

• Run hour equalisation: the BMS rotates the active/standby designation between CRAC units on a programmed schedule (typically weekly), ensuring all units accumulate similar run hours and that standby units are regularly exercised.
• Automatic failover: if an active CRAC unit faults, the BMS starts the standby unit immediately — without requiring manual intervention. The transition must complete before room temperature begins to rise towards alarm thresholds.
• Cascade sequencing: as IT load varies, the BMS adds or removes active CRAC units to maintain supply temperature while keeping active units at optimal part-load efficiency.

How does the BMS handle monitoring, alerting, and DCIM integration?

Data centre operations teams require continuous visibility of their environment. ISO/IEC 27001 certification — increasingly required for colocation operators and cloud service providers — includes physical environmental controls as a mandatory control domain, with BMS monitoring records forming part of the evidence base. Any BMS that is network-connected or remotely accessible must also be evaluated for cybersecurity risk — for a guide to BMS network threats and how to address them, see our article on BMS cybersecurity. The BMS provides operational visibility through:

• 24/7 alarm monitoring: all critical alarms are monitored continuously. Alpha Controls offers maintenance contracts with 24/7 alarm response and guaranteed engineer attendance times.
• Escalation policies: alarms escalate through a defined chain — BMS system to on-site engineer to on-call specialist — with automatic escalation if acknowledgement is not received within defined timeframes.
• DCIM integration: Data Centre Infrastructure Management platforms (such as Nlyte, Sunbird, or Vertiv TRELLIS) aggregate BMS data alongside asset management, capacity planning, and power data. The BMS exposes data via REST API, SNMP, or BACnet/IP to feed these platforms, providing a single pane of glass for data centre operations.
• Trend data for capacity planning: BMS trend logs of temperature, humidity, power, and PUE over time provide the data needed to plan future capacity expansions and identify efficiency improvement opportunities.

How do you choose a contractor for a data centre BMS project?

Data centre BMS projects require a contractor with specific experience in critical infrastructure — not just general commercial building controls. Alpha Controls brings expertise in CRAC unit protocol integration, precision cooling control, leak detection systems, and the alarm management strategies that critical facilities demand.

We work on colocation data centres, enterprise server rooms, and edge computing facilities across London and the South East. Our networking services complement BMS work in facilities where structured cabling and BMS infrastructure are being installed together.

Contact Alpha Controls to discuss your data centre BMS project, or explore our BMS services and leak detection pages for more information.