CDCP key terms

A four-level scheme describing data centre availability and fault tolerance, from Tier I (basic capacity, no redundancy) to Tier IV (fully fault tolerant with concurrently maintainable critical paths).

A design property allowing any piece of critical equipment to be removed for maintenance or repair without shutting down the IT load, typically associated with Tier III and IV topologies.

The ability of a facility to sustain an unplanned failure of any single component in the critical path without interrupting the IT load, achieved through redundant capacity and automatic failover.

Spare capacity provided beyond the minimum needed to carry the load. N is the minimum, N+1 adds one spare unit, and 2N duplicates the entire capacity for full backup.

Any component whose failure would interrupt the entire IT load because no alternate path or spare capacity exists to carry the load during its outage.

The sequence of power, cooling, and control components that must function for IT equipment to remain powered and cooled, from utility entry through to the rack.

A widely used data centre infrastructure standard covering topology, cabling, power, cooling, and security requirements, with an accompanying tier rating framework and annexes for related facilities.

The ratio of total facility power to IT equipment power. A value of 1.0 is ideal; higher values indicate more energy consumed by cooling, losses, and other support systems.

The reciprocal of PUE, expressed as a percentage, representing the share of total facility power that actually reaches IT equipment; equal to 1 divided by PUE.

The range of inlet temperature and humidity conditions within which IT equipment is permitted to operate, defined by ASHRAE classes A1 through A4 and used in cooling design.

A standardized hierarchical cabling system using defined media, connectors, and pathways, organized into backbone and horizontal distribution to support reliable, manageable data centre connectivity.

A patching point where cables terminate and are interconnected, used in main distribution, horizontal distribution, or equipment areas to manage and reconfigure connections flexibly.

A controlled space where external carriers and service providers terminate their networks and interconnect with the data centre's internal cabling infrastructure.

The point where electrical power from the utility grid enters the facility, including transformers, switchgear, and metering that define the site's available capacity and reliability.

A system that provides conditioned, continuous power to IT loads during brief outages and transfers to generator or utility power, bridging the gap without interruption.

A UPS topology that converts incoming AC to DC for the battery and back to clean AC for the load, isolating equipment from sags, surges, and frequency variations at all times.

A series-connected group of batteries providing DC energy to a UPS during outages; its runtime, age, temperature, and maintenance determine the ride-through available before generator power is ready.

A device that senses loss of the primary power source and automatically switches the load to an alternate source, such as a generator, within a defined transfer time.

An engine-driven alternator that supplies backup electrical power when utility power fails, characterized by its rated capacity, fuel supply, startup time, and load acceptance capability.

A resistive or reactive device used to apply controlled electrical load to a generator or UPS for testing, commissioning, and verifying capacity without risking live IT equipment.

Equipment that distributes electrical power to racks and IT devices; floor-mounted PDUs step down voltage and feed branch circuits, while rack PDUs provide outlet-level distribution.

A prefabricated, modular electrical distribution system using copper or aluminum conductors in a housing, allowing tap-off boxes to be added or moved above racks without recabling.

The final circuit from a panelboard or PDU to the load, protected by an overcurrent device sized to the conductor ampacity and the connected equipment requirements.

The electrical load a rack position is designed to support, used to plan cooling, power distribution, and floor loading; high-density racks may exceed 15–20 kW each.

The ratio of real power (kW) to apparent power (kVA) in an AC circuit; low power factor increases current for the same useful work and affects UPS and generator sizing.

An arrangement where racks face each other in alternating rows so cold supply air enters rack fronts and hot exhaust air is collected from rack rears, reducing mixing and improving cooling efficiency.

Physical barriers such as doors, panels, and roofs that enclose either the cold aisle or hot aisle to prevent air mixing between supply and return streams, improving cooling capacity and efficiency.

Computer room air conditioning (direct expansion) or air handling (chilled water) units that cool and circulate air in the data hall; CRAH units use chilled water coils and typically lack built-in compressors.

A refrigeration machine that removes heat from water in a chilled water loop serving CRAH units or other cooling equipment; its efficiency and redundancy strongly affect overall facility PUE.

A cooling strategy that uses cool outdoor air or water instead of mechanical refrigeration when ambient conditions allow, reducing compressor energy and lowering PUE.

The space between a structural slab and a raised access floor, often used to deliver conditioned air to perforated tiles or to route cabling and piping beneath the data hall.

The temperature of air delivered to IT equipment inlets, a key control setpoint balancing equipment reliability against cooling energy use and condensation risk.

Managing moisture in the data hall to avoid electrostatic discharge at low humidity and condensation or corrosion at high humidity, typically through humidification and dehumidification equipment.

Cooling IT equipment by circulating liquid directly to or near heat sources, such as cold plates or immersion tanks, enabling higher rack power densities than air cooling alone.

Numerical modeling of airflow, temperature, and pressure in a data hall, used to predict hot spots, evaluate containment options, and validate cooling designs before construction.

A scheduled period during which equipment may be taken out of service for testing, repair, or upgrades, planned to avoid impact on IT loads in concurrently maintainable designs.

A controlled process for reviewing, approving, scheduling, and documenting changes to infrastructure or IT systems, reducing the risk of unplanned outages from undocumented modifications.

A detailed, step-by-step document describing how to perform a specific task safely, including prerequisites, verification steps, rollback plans, and approvals required before execution.

A documented routine procedure for operating equipment consistently and safely, used for recurring activities such as transfers, startups, shutdowns, and inspections.

A predefined response procedure for emergencies such as power loss, fire, or cooling failure, specifying actions, responsibilities, and escalation to protect people and IT loads.

The process of detecting, logging, responding to, and resolving service disruptions, including root cause analysis and corrective actions to prevent recurrence.