Data centres

Operational Visibility for Cooling, Power and Critical Infrastructure

Data centre facilities engineering is unforgiving: the consequences of a cooling or power failure are immediate, and the evidence requirements afterwards are absolute.

Request a Demo See the Platform

The business problem

Specialist systems, generalist maintenance records

Cooling and power infrastructure is monitored in detail by specialist systems, but the maintenance, contractor attendance and change records around it are often kept to a far lower standard — in spreadsheets, email threads and contractor PDFs. When an incident is reviewed, the engineering data exists and the operational record does not.

Detailed monitoring alongside weak maintenance recordsContractor attendance evidenced by emailed reportsChange and work history not reconstructable per assetRedundancy status assessed manually

The IntelliGrid approach

An FM operational layer above specialist monitoring

IntelliGrid integrates with existing monitoring and DCIM rather than replacing it, and supplies the operational discipline around it: asset-level maintenance history, permit-controlled contractor access, evidenced work and full audit trail — with alarm correlation to reduce noise.

Integration alongside existing DCIM and monitoringAsset-level maintenance and change historyPermit-controlled, time-bound contractor accessAlarm correlation and owned incident handling

AI use cases

CRAC thermal anomalies

Capacity loss detected from supply and return temperature behaviour before rack impact.

UPS battery condition

Anomalies in battery and charging behaviour flagged between tests.

Alarm correlation

Related alarms grouped so genuine incidents surface from volume.

Sustainability opportunities

Cooling efficiency

Plant operation and setpoint strategy assessed against IT load.

Energy intensity

Facility consumption tracked against IT load for efficiency reporting.

Water use in cooling

Consumption monitored where evaporative cooling is used.

Challenges

What makes this environment demanding

Cooling

CRAC and chilled water systems where capacity loss escalates within minutes.

UPS and power

Uninterruptible supply, switchgear and distribution under continuous duty.

Generators

Standby readiness verified between statutory tests.

Temperature and humidity

Tight envelopes monitored at rack and room level.

Leak detection

Water in a data hall requiring immediate, procedural response.

Alarm volume

Large numbers of alarms where genuine events must not be lost.

Redundancy monitoring

N+1 or 2N status and the effect of any component out of service.

Change control

Every physical intervention documented and attributable.

Workflow

Controlled intervention in a live hall

STEP 1

Raise

Work identified from monitoring, predictive risk or a PPM plan.

STEP 2

Assess

Redundancy impact and risk to the hall evaluated.

STEP 3

Permit

Change approved with a defined window and method.

STEP 4

Access

Time-bound access issued to the named engineer.

STEP 5

Evidence

Work, readings and sign-off recorded against the asset.

Data centre facility KPIs

Metrics reported by the platform. Values depend on your operation.

Critical plant uptime

Cooling and power availability

Change compliance

Interventions performed under permit

PPM compliance

Statutory and manufacturer routines

Open high-risk items

Current predictive exposure

Business benefits

Maintenance records matching the standard of the monitoring Every physical intervention permitted and attributable Alarm noise reduced through correlation Contractor access time-bound to approved work Incident review supported by a complete operational timeline

Connects with

BMS DCIM CRAC units Chiller plant UPS and switchgear Generators Leak detection Environmental sensors Access control

See integration architecture →

Frequently asked

Does IntelliGrid replace DCIM?

No. Specialist DCIM and monitoring platforms remain in place. IntelliGrid operates as the facilities management layer above them — maintenance, work orders, contractor control, permits and audit trail — and integrates with those systems rather than duplicating their function.