Presight and Khazna to Modernize Data Centre Operations with Applied Intelligence
Presight has signed an MOU with Khazna Data Centers to deploy an AI-driven command-and-control platform across 30 data centres in the UAE. The system will run from a secure AI hub in Abu Dhabi and bring unified oversight to energy, cooling, equipment performance, and security. The objective is simple: predict issues before they occur and keep facilities operating efficiently, 24/7.
Why this matters for operations
For ops leaders, this is a move from reactive to anticipatory management at portfolio scale. With standardized telemetry and machine learning in the loop, the platform can cut downtime, improve resource usage, and tighten SLA performance across sites.
- Early fault detection to reduce incidents and mean time to repair (MTTR).
- Adaptive cooling strategies to lower energy consumption and improve PUE.
- Predictive maintenance to prioritize work orders and avoid unnecessary servicing.
- Unified 24/7 oversight to coordinate response and ensure continuity across all locations.
How the platform is expected to operate
The command-and-control centre will ingest real-time data from all Khazna facilities, apply analytics to forecast risk, and orchestrate actions through existing BMS, DCIM, and CMMS integrations. This includes policy-driven automation for setpoints, maintenance scheduling, and security workflows.
The centre will also sit within G42's Intelligence Grid, connecting UAE operations with a growing global footprint spanning Singapore, Kazakhstan, Europe, and Africa. That linkage enables standardized governance, shared models, and consistent playbooks across regions.
What leaders should prepare now
- Consolidate your asset inventory and CMDB; map sensors, controllers, and data schemas by site.
- Audit integrations: BMS/DCIM/CMMS, ticketing, SIEM/SOC, and relevant OT networks; confirm API access and rate limits.
- Define target KPIs: PUE/WUE, MTBF, MTTR, incident rate, energy per MW of IT load, and SLA thresholds by tier.
- Set alert policies and escalation paths; standardize runbooks for the top 10 failure modes (cooling, power, network, security).
- Establish change control for autonomous actions (who approves which automations, rollback rules, audit trails).
- Validate cyber posture for OT/IT convergence: network segmentation, credential vaulting, logging, and anomaly detection.
- Schedule tabletop exercises and simulated faults to test decision logic before full-scale activation.
- Plan workforce enablement: operator training, on-call rotations, and clear ownership across facilities.
Impact you can quantify
- Downtime: fewer surprise incidents and faster restoration through early fault detection.
- Energy: improved cooling efficiency and setpoint control to pull PUE downward.
- Maintenance: shift from calendar-based to condition-based schedules to reduce unnecessary truck rolls.
- Security and compliance: centralized visibility, standardized reporting, and auditable actions.
Leadership perspective
Presight's CEO, Thomas Pramotedham, emphasized embedding AI at the core of mission-critical systems to drive efficiency, resilience, and sustainability at scale-advancing the UAE's digital ambitions with high-standards facility operations.
Khazna's CEO, Hassan Alnaqbi, underscored the shift to intelligent, adaptive infrastructure that anticipates and optimizes in real time, setting a strong benchmark for sustainable data centre management.
Bottom line for ops teams
This initiative sets a clear operating model: centralize data, standardize playbooks, and let intelligent systems handle the noise while people focus on high-impact decisions. If you prepare your data, integrations, and governance now, you can capture uptime gains and energy savings quickly as the rollout progresses.
Upskill your team
If you're building capabilities for AI-driven operations-automation, monitoring, and incident response-explore practical training paths here: AI courses by job role.
Your membership also unlocks:
