Ensuring reliable power supply is crucial for the continuous operation of data centres. This checklist covers the critical aspects of power systems, including utility provisions, UPS, batteries, and generators. By following this guide, data centre operators can proactively identify risks, implement best practices, and maintain uninterrupted services.

---

Power Continuity in Data Centres | Ensuring Uninterrupted Operations

Power continuity is one of the most critical aspects of data centre design and operations. A robust and reliable power infrastructure ensures that mission-critical systems remain operational even in the face of utility failures, natural disasters, or equipment malfunctions. Data centres achieve this through a combination of multiple utility power feeds, uninterruptible power supplies (UPS), and generators, each playing a distinct role in the power continuity chain.

---

• Multiple Utility Power Feeds

Description:
Data centres often rely on multiple power feeds from independent utility grids to ensure redundancy. These feeds are sourced from geographically or electrically diverse substations to mitigate the impact of a failure in a single grid.

How It Works:

• Each utility feed provides a separate path for power delivery to the data centre.

• Redundant feeds are connected through Automatic Transfer Switches (ATS), which detect power loss and switch seamlessly to the alternative feed without interruption.

Benefits:

• Reduces dependency on a single utility provider, improving resilience against regional outages.

• Ensures continuous power supply in the event of equipment failure or maintenance on one feed.

• Offers protection from localised faults such as transformer failures or distribution issues.

Limitations:

• Utility feeds alone cannot protect against widespread grid failures, such as during extreme weather events or rolling blackouts.

---

• Uninterruptible Power Supplies (UPS)

Description:
A UPS system acts as an immediate buffer between the data centre and its power sources, providing temporary power in case of interruptions. Modern UPS systems use batteries or flywheels to supply power for short durations.

How It Works:

• During normal operations, the UPS conditions incoming power to eliminate spikes, surges, or sags.

• If utility power fails, the UPS instantly takes over, ensuring zero downtime for critical systems.

• The UPS provides sufficient power for systems to continue operating until backup generators activate.

Benefits:

• Zero transition time ensures uninterrupted power delivery to IT equipment.

• Protects against both short-term outages and power quality issues, such as voltage irregularities.

• Serves as a bridge to longer-term solutions, like generators.

Limitations:

• Limited runtime based on battery or flywheel capacity, typically minutes rather than hours.

• Requires regular maintenance and testing to ensure reliability.

---

• Backup Generators

Description:
Generators provide long-term backup power by converting fuel (diesel or natural gas) into electricity. These systems activate automatically when utility power fails.

How It Works:

• Upon detecting a utility failure, the generator starts and synchronises with the data centre's power systems, usually within seconds.

• Generators are designed to operate for extended periods, limited only by fuel availability and system maintenance.

Benefits:

• Provides reliable, long-duration backup power, essential during prolonged outages.

• Scalable to support the full load of the data centre, including cooling and auxiliary systems.

• Typically configured in N+1 or 2N redundancy to ensure availability even during maintenance or generator failure.

Limitations:

• Dependence on fuel availability and logistics for refuelling during extended outages.

• Higher operational costs due to fuel consumption and periodic maintenance.

---

The Power Continuity Chain in Action

The synergy between these components ensures seamless power continuity:

• Utility Power as Primary Source:
  Multiple feeds provide redundancy to minimise the risk of utility failures.

• UPS as Instantaneous Backup:
  When utility power is lost, the UPS provides an immediate power source, protecting IT equipment from abrupt shutdowns.

• Generators for Extended Backup:
  While the UPS handles the transition, the backup generators activate, taking over the load for as long as needed.

---

Best Practices for Power Continuity

• Redundancy Planning: Implement N+1 or 2N configurations for UPS and generators to avoid single points of failure.

• Regular Testing: Schedule routine load tests for generators and battery discharge tests for UPS to ensure readiness during emergencies.

• Fuel Management: Maintain sufficient on-site fuel reserves and establish contracts with fuel suppliers for rapid replenishment.

• Monitoring and Maintenance: Deploy advanced monitoring tools to identify power anomalies and schedule predictive maintenance.

---

By integrating multiple utility feeds, UPS systems, and backup generators, data centres achieve unparalleled power reliability. This layered approach safeguards critical infrastructure, ensuring uninterrupted operations in an increasingly digital world.

---

Power Checklist for Data Centres

1. Critical Equipment Classification

• What is classified as critical equipment? Ensure power availability for critical servers, switches, routers, emergency lighting, access control, surveillance, building management, and telephony systems.

• Redundancy Planning: Identify dependencies and redundancy measures for critical infrastructure to minimise disruptions.

---

2. Utility Power Provision

• Current Capacity: Assess the municipal/utility power provision in kVA, kW, or MW.

• Data Centre Power Usage: Monitor total power consumption, including critical and non-critical loads.

• Maximum Capability: Determine the maximum power capacity for all activities and for the UPS alone.

• Municipal Growth Capability: Confirm whether the local municipality can meet future power demand.

• Power Stability: Ensure the municipal/utility power supply is stable, with mitigation plans for instability.

• Cable Location: Identify whether power supply cables are overhead or underground to assess vulnerability.

---

3. Redundant Power Configurations

• Dual Power Feeds: Verify if 100% redundant configurations (e.g., A/B feeds) are implemented.

• Separate Grids: Ensure power feeds originate from separate municipal grids to reduce simultaneous failure risks.

---

4. Emergency Power Off (EPO)

• Installation: Confirm the presence of EPO and clearly mark its location for emergencies.

• Functionality: Regularly test EPO to ensure it can isolate power during critical incidents.

---

5. UPS Systems

• Inventory and Capacity: Record the number, age, and capacity of UPS units, including feed A and feed B.

• Ownership: Clarify whether UPS units are self-owned, rented, or leased.

• Maintenance: Maintain and test schedules for each UPS, with logs of the latest tests.

• Auto-Bypass: Ensure an auto-bypass system transfers loads to raw power during UPS failure.

• Manual Bypass: Verify the availability of manual bypass mechanisms for maintenance purposes.

• Location: Determine if UPS units are located inside or outside the computer room to manage heat and noise.

• Load Balancing: Apply regulations to balance UPS loads within 5%-10% to optimise performance.

---

6. Battery Systems

• Location and Safety: Batteries should be housed in separate, ventilated, and environmentally controlled rooms with blast screens.

• Specifications: Document the number of battery strings, capacity (in Ah), voltage per block, and autonomy time under load.

• Condition Monitoring: Regularly inspect battery terminals, cables, and blocks for wear or damage.

• Life Expectancy: Measure remaining battery life and ensure testing intervals align with industry standards.

• Battery Fusing: Verify that battery strings are individually fused for safety.

---

7. Generators

• Quantity and Capacity: List the number of generators installed and their capacities (continuous and normal ratings).

• Start-Up Time: Monitor generator start-up times to minimise service interruptions during outages.

• Maintenance: Adhere to regular testing intervals for free-run, linear, and non-linear loads. Maintain test logs.

• Fuel Management: Track day and bulk tank capacities, test fuel for contamination annually, and maintain SLA agreements with fuel suppliers.

• Leak Detection: Install and connect fuel leak detection systems to the BMS/EMS.

• Paralleling Switch Gear: Ensure the system supports automated paralleling for load sharing and redundancy.

---

8. Automatic Transfer Switches (ATS)

• Inspection: Schedule periodic inspections and thermal scans for all ATS devices.

• Condition Monitoring: Assess the condition of ATS systems and ensure uninterrupted transition during power failures.

---

9. Monitoring & Automation

• Building Management Integration: Connect all critical power systems, including UPS, generators, and fuel tanks, to the Building Management System (BMS) for centralised monitoring.

• Alarms and Notifications: Implement automated alerts for low fuel levels, power fluctuations, and equipment faults.

• Thermal Scanning: Conduct thermal imaging regularly for all electrical infrastructure to detect overheating or imbalances.

---

10. Documentation & Compliance

• Maintenance Reports: Ensure all maintenance and test reports are up-to-date and available for review.

• Compliance Standards: Align power infrastructure with Tier III or Tier IV design standards, if applicable.

---

Wrap

Power reliability in a data centre is a critical aspect of operations, demanding regular assessments and adherence to industry best practices. This checklist is designed to empower data centre operators to proactively address vulnerabilities, optimise power systems, and ensure seamless operations, even during utility outages or equipment failures.

---

Additional reading: