DCAT key terms

A standard measure of vertical height in a server rack, where 1U equals 1.75 inches (44.45 mm); equipment is sized in U to ensure it fits and aligns with mounting rails.

A diagram showing the front view of a rack with each piece of equipment placed at specific U positions, used for planning space, power, and cable routing.

A contained or open aisle where cold supply air is delivered to the front of servers; separating it from hot exhaust air improves cooling efficiency.

The aisle that receives warm exhaust air from the rear of servers; containing it prevents hot air from mixing with cold supply air.

The practice of arranging equipment, blanking panels, and containment to ensure cold air reaches server intakes and hot air returns to cooling units without mixing.

A metal or plastic cover that fills unused rack units to prevent hot exhaust air from recirculating to the cold aisle through empty spaces.

A second power supply installed in a server or device so that if one fails, the other continues to provide power, often connected to separate power feeds.

A device that distributes electrical power to multiple outlets in a rack; a metered or switched PDU provides monitoring and remote control of individual outlets.

A battery-backed device that provides temporary power during an outage or voltage fluctuation, allowing equipment to ride through short interruptions or shut down gracefully.

A backup power source that produces electricity from fuel (often diesel or natural gas) to sustain data center operations during extended utility outages.

A device that automatically switches the electrical load from the primary utility feed to a backup generator when it detects a power failure.

Computer Room Air Conditioner: a cooling unit that regulates temperature and humidity in a data center by blowing cold air into the room and drawing warm air back.

Computer Room Air Handler: a cooling unit that uses chilled water instead of refrigerant to cool air, often paired with a chiller plant for large data centers.

A large cooling system that removes heat from water (or another fluid) and rejects it outdoors, supplying chilled water to CRAH units or other cooling equipment.

A structure that rejects heat from a chiller or industrial process by evaporating water, releasing waste heat to the atmosphere.

A computer that provides services or resources to other computers over a network, typically housed in a data center and designed for continuous operation.

A modular server that fits into a chassis along with other blades, sharing power, cooling, and networking components to save space and reduce cabling.

A type of memory module used in servers, with pins on both sides of the circuit board that connect independently to the motherboard.

A high-performance storage protocol designed for solid-state drives, using the PCIe bus to deliver lower latency and higher throughput than older interfaces.

A method of combining multiple physical drives into a logical unit to improve performance, provide fault tolerance, or both, depending on the RAID level.

A standby drive that is powered on and ready to automatically replace a failed drive in a RAID array without manual intervention.

A storage device that uses rotating magnetic platters to read and write data, offering high capacity at lower cost but with slower access than solid-state drives.

A storage device that uses flash memory with no moving parts, providing faster access times and greater shock resistance than hard disk drives.

A network switch installed at the top of a rack that connects servers within that rack to the data center network, reducing cable runs.

A connection from an access switch (such as a ToR switch) to a higher-level switch or network, carrying traffic from many servers toward the core.

A cable that transmits data as pulses of light through glass or plastic strands, supporting high bandwidth and long distances with immunity to electromagnetic interference.

A cable made of pairs of insulated copper wires twisted together to reduce interference, commonly used for short-distance Ethernet connections in data centers.

A standardized layout of cables, patch panels, and connectors that supports a data center's network and simplifies moves, adds, and changes.

A hardware panel with multiple ports that terminates incoming cables and allows flexible connections to network equipment via short patch cords.

The practice of organizing and routing cables using trays, ties, and labels to maintain airflow, simplify maintenance, and reduce accidental disconnections.

The ability to remove and replace a component (such as a drive or power supply) while the system remains powered on and operational.

The sudden flow of static electricity between objects, which can damage sensitive electronic components; prevented by using wrist straps and grounded mats.

A device worn on the wrist that safely discharges static electricity from the body to a grounded point, protecting components during handling.

A separate network path that allows administrators to monitor and control servers independently of the main data network, often via a dedicated management port.

A specialized microcontroller on a server motherboard that provides out-of-band management, enabling remote power control, console access, and hardware monitoring.

A standard protocol for monitoring and managing network devices, allowing collection of performance data and sending of alerts (traps) to a management system.

A standard protocol and format for sending event log messages from devices to a central logging server for analysis and retention.

A safety procedure that isolates energy sources and applies locks and tags to prevent accidental energization during maintenance or repair.

Gear such as gloves, eye protection, and arc-rated clothing worn to protect workers from hazards including electrical shock, burns, and physical injury.

A cabinet that houses circuit breakers and distributes electrical power to circuits; in a data center, panels feed PDUs and must be clearly labeled and accessible.