Learn · Glossary · 62 core terms

Data center glossary

Plain-English definitions of the terms you'll hear in job ads, interviews, and training. 62 core terms, grouped by the system they belong to.

Get 25 Free Questions Get the Question Bank

Power & electrical

How a facility keeps clean, continuous power flowing to the equipment.

UPS Uninterruptible Power Supply: A battery-backed system that carries the electrical load for a short time when utility power drops, holding equipment up until generators start or power returns.; Trace it in the free power train explorer; Appears in: Critical Facilities Technician role guide · MEP Design Engineer role guide · Commissioning Engineer role guide · taught in Fundamentals
PDU Power Distribution Unit: Equipment that takes incoming power and splits it out to racks and devices. A floor or rack PDU distributes and, in many cases, meters the power each rack draws.; Trace it in the free power train explorer; Appears in: Data Center Technician role guide · taught in Fundamentals
ATS Automatic Transfer Switch: A switch that automatically moves the load from one power source to another, for example from utility to generator, without someone flipping it by hand.; Trace it in the free power train explorer; Appears in: Critical Facilities Technician role guide · Commissioning Engineer role guide · Controls / BMS Engineer role guide · taught in Fundamentals
Generator: An engine-driven backup power source, usually diesel, that supplies the site during an extended utility outage after the UPS bridges the first seconds or minutes.; Trace it in the free power train explorer
Busway: An overhead track of enclosed conductors that distributes power along a row of racks, letting technicians tap off power where they need it instead of running individual whips.; Trace it in the free power train explorer

Cooling & environment

Keeping temperature and airflow inside the safe range for the hardware.

Cold supply (solid) rises through the raised floor into the contained cold aisle; hot exhaust (dashed) returns above the racks to the CRAC. Airflow separation is the real job of data center cooling.

CRAC Computer Room Air Conditioner: A room cooling unit that uses a refrigerant compressor to remove heat from the data hall, much like a large, precise air conditioner built for continuous duty.; The distinction that matters Tell them apart by what arrives at the unit. A CRAC has its own compressor and refrigerant, so it makes cold locally. A CRAH has chilled water pipes and fans, and the cold was made centrally by a chiller. If someone asks what happens when the chiller trips, the CRAH stops cooling and the CRAC does not, and that is usually the real question.
CRAH Computer Room Air Handler: A room cooling unit that uses chilled water and fans instead of its own compressor. The cooling is produced centrally by chillers and delivered as cold water.; The distinction that matters A CRAH has no compressor of its own, so it is only as good as the chilled water loop feeding it. That makes the chiller plant, the pumps and the loop temperature part of your answer whenever a CRAH is involved, which is not true of a CRAC.; Appears in: Critical Facilities Technician role guide · Controls / BMS Engineer role guide · taught in Fundamentals
Hot aisle / cold aisle: A rack layout where cold supply air and hot exhaust air are kept on separate aisles so they don't mix, which makes cooling far more efficient.
Containment: Physical barriers (doors, panels, or roofs over an aisle) that seal hot and cold air apart so the cooling system doesn't waste energy mixing them.
Chiller: Plant equipment that produces chilled water for CRAH units and other cooling loops by rejecting heat to the outside, often the largest single energy user on a site.
PUE Power Usage Effectiveness: A ratio that describes how much of a facility's total electrical draw actually reaches the IT equipment, calculated as total facility power divided by IT power. A result of 1.0 would mean every watt reached the servers, which is not achievable in practice; well-run sites typically report somewhere between 1.2 and 1.5. Anyone quoting 1.0 has misunderstood the measure.
Economizer Free cooling: A cooling mode that uses outside conditions instead of mechanical refrigeration when the weather allows, either by drawing in cool outside air or by rejecting heat through a water loop. It reduces the energy the chillers and compressors have to spend, so sites in cooler climates run it for a large part of the year and report lower efficiency ratios as a result.

Racks & physical space

How equipment is mounted, measured, and organized on the floor.

Rack unit (U): The standard height measurement for rack-mounted equipment. One U is 1.75 inches. A "1U server" takes a single slot; a full rack is commonly 42U or 48U.; Appears in: Data Center Technician role guide · taught in Fundamentals
Cabinet / rack: The vertical frame or enclosure that servers, switches, and cabling mount into. Cabinets are usually enclosed with doors; open racks are frames without side panels.
Remote hands: On-site technician work performed on behalf of a customer or remote team, swapping a drive, reseating a cable, power-cycling a device, so they don't have to travel to the site.; Appears in: Data Center Technician role guide · taught in Fundamentals
Blanking panel: A solid filler plate that covers empty rack slots so cold supply air can't short-circuit through gaps to the hot side, protecting airflow separation.
Colocation Colo: A model where an operator provides the building, power, cooling and connectivity, and customers install and own the equipment in space they rent, measured in cabinets, cages or power drawn.; The distinction that matters In a colo, the hardware in the rack usually belongs to someone else, which changes what you are allowed to touch and who must approve it. That is why remote hands, cross-connects and access escorting are core colo work and barely feature at a single-tenant site.
Hyperscaler: An operator running data centers at very large scale for its own platform, typically with standardized designs repeated across many sites and its own operating procedures.; The distinction that matters The practical difference for a candidate is standardization. Hyperscale work tends to be deep in a narrow, heavily proceduralized scope across identical sites; colocation work tends to be broader and more varied per site because every customer is different. Neither is the junior option, and interviewers notice when you know which one you are applying to.

Cabling & network

The physical layer that connects everything and keeps it traceable.

Structured cabling: A planned, standardized cabling system with defined pathways, labeling, and connection points, so that connections are consistent, documented, and easy to trace.; Appears in: Data Center Technician role guide · Network & Cabling Technician role guide · taught in Fundamentals
Patch panel: A mounted panel of ports that terminates permanent cabling, letting technicians make and change connections with short patch cords instead of re-running fixed cable.; Appears in: Network & Cabling Technician role guide · taught in Fundamentals
Cross-connect: A managed physical connection between two parties or systems, usually made at a patch panel or meet-me area, used heavily in colocation to link customers and carriers.; The distinction that matters A patch panel is hardware you can point at. A cross-connect is a managed connection, usually recorded in a customer's contract and in the operator's records, which is why moving one is paperwork before it is cable work. Treating a cross-connect as just a patch is how an unplanned customer outage happens.; Appears in: Network & Cabling Technician role guide · taught in Fundamentals
Single-mode / multimode fiber: Two fiber-optic cable types. Single-mode carries light over long distances with a narrow core; multimode suits shorter runs inside a facility with a wider core.; Appears in: Network & Cabling Technician role guide · taught in Fundamentals
MDA / HDA Main / Horizontal Distribution Area: Standard cabling zones in a data center. The MDA is the central connection point; HDAs distribute out toward the equipment rows.; Appears in: Network & Cabling Technician role guide · taught in Fundamentals

Redundancy & reliability

How sites stay running when a component or a whole path fails.

N+1: A redundancy level where there is one spare unit beyond what the load needs. If "N" units are required, the site runs N+1 so any single unit can fail or be serviced without impact.; The distinction that matters N+1 is a redundancy count, not a reliability rating and not a tier. It says how many spare units exist; it does not say whether the paths are independent, which is what concurrent maintainability and fault tolerance describe. Two sites can both be N+1 and behave completely differently in a failure. Also worth saying out loud: while one unit is out for service, the spare is spent and the site is running at N.; See N, N+1 and 2N applied in the free explorer; Appears in: Critical Facilities Technician role guide · MEP Design Engineer role guide · Commissioning Engineer role guide · taught in Fundamentals
2N: A redundancy level with two complete, independent systems, full duplication. Either side can carry the entire load, so an entire path can fail without downtime.; The distinction that matters 2N duplicates the whole path, not just the unit, which is what separates it from N+1. If both halves share a single downstream component, the site is not really 2N there, and finding that shared component is the single-point-of-failure question in disguise.; See N, N+1 and 2N applied in the free explorer; Appears in: MEP Design Engineer role guide · taught in Fundamentals
Redundancy: Designing spare capacity or duplicate paths into power, cooling, and network so that a single failure doesn't take the load down.
Single point of failure SPOF: Any one component whose failure would bring down a service because nothing backs it up. Reliable designs work to identify and eliminate SPOFs.; Appears in: MEP Design Engineer role guide · taught in Fundamentals
Concurrent maintainability: A design property where any single component can be taken offline for maintenance while the site keeps running the load at full function.; Appears in: MEP Design Engineer role guide · taught in Fundamentals
Uptime Tier Tier I to Tier IV: A four-level classification of data center infrastructure topology, written in Roman numerals. Tier III is concurrently maintainable, meaning any single component can be taken out for service without shutting the load down. Tier IV is fault tolerant, meaning the site survives an unplanned failure of any single component. A listing that writes it in Arabic numerals is usually describing an aspiration rather than a certified rating.; The distinction that matters A tier describes the whole site topology as assessed by a certifying body. N+1 and 2N describe the redundancy of one system. A site can run N+1 cooling and not be Tier III, and quoting a redundancy level when you were asked for a tier is the most common vocabulary slip in a facilities screen.

Operations & safety

The day-to-day process, response, and safety language of the floor.

Runbook: A documented set of steps for handling a specific task or situation, so that any qualified technician can carry it out consistently and safely.; The distinction that matters A runbook is the standing instruction for a recurring task or situation. A MOP is written for one specific piece of work on live infrastructure and is approved before it is used. If you are asked which you would follow, the honest answer names the site's own rule and does not treat the two as interchangeable.; Appears in: Data Center Technician role guide · NOC Operator role guide · taught in Fundamentals
Incident: An unplanned event that disrupts or threatens a service. Data center operations track, escalate, and resolve incidents against defined response targets.; Appears in: NOC Operator role guide · taught in Fundamentals
Escalation path: The defined order in which an issue is raised to more senior or specialized people when it can't be resolved at the current level or within a time limit.; Appears in: NOC Operator role guide · taught in Fundamentals
SLA Service Level Agreement: A commitment to a measurable level of service, such as uptime or response time, that operations teams are expected to meet and report against.; Appears in: NOC Operator role guide · taught in Fundamentals
Change management: The controlled process for planning, approving, and recording changes to live systems so that risky work is reviewed and reversible rather than ad hoc.; The distinction that matters Change management decides whether and when work is allowed; a MOP describes how that approved work is carried out. Naming both, in that order, is what separates a candidate who has worked under change control from one who has read about it.
LOTO Lockout / Tagout: A safety procedure that isolates energy sources and locks them off before work, preventing equipment from being powered on while someone is working on it.; Appears in: Critical Facilities Technician role guide · Trades on Site role guide · taught in Fundamentals
BMS Building Management System: The control system that monitors and operates the mechanical and electrical plant: cooling units, pumps, valves, generators and the sensors feeding them. It holds the setpoints, runs the sequences that start and stop equipment, and raises the alarms an operator sees first. In an interview, being able to say what the BMS controls versus what it only reports is the distinction that matters.
DCIM Data Center Infrastructure Management: Software that tracks the physical estate: what is installed in which rack, how much power and space each cabinet is drawing, where circuits and cables run, and how capacity is trending. It answers questions about the floor rather than about the plant, which is what separates it from a building management system, and it is the record technicians update after every install or decommission.
MOP Method of Procedure: A written, approved, step-by-step script for carrying out a specific piece of work on live infrastructure. A good one states the prerequisites, numbers each action simply, puts a verification point after every critical step, and defines the abort criteria that send the team back to a safe state. Work on critical systems is expected to follow an approved MOP rather than improvisation.; The distinction that matters A MOP is approved for one job at one time. A runbook stands until it is revised. Change management is the process that decides whether the work happens at all. An interviewer asking about a MOP is usually checking whether you would improvise when the script does not fit reality, and the safe answer is to stop at the abort criteria rather than to adapt on the floor.
Severity level Sev / P1 to P4: The rating that says how bad an incident is and therefore how fast and how far it must be escalated. Most sites run a small numbered scale where the top level means live service is affected and the bottom means a fault with no customer impact yet. The exact definitions, response times and escalation contacts are set by each employer.; The distinction that matters Severity is impact, priority is the order you work things in, and the two are set by different rules. Assigning a severity is not a judgment call you make alone: sites publish the matrix precisely so that the same event gets the same rating at three in the morning as it does at midday. In an interview, say you would rate it against the site's matrix and escalate rather than negotiate the number down.; Appears in: NOC Operator role guide · taught in Fundamentals
EPO Emergency Power Off: A dedicated system that removes electrical power from a room or space in an emergency, operated from clearly marked stations. It exists for life safety, and once operated it drops the load it covers.; The distinction that matters An EPO is not a power switch and it is not a test you volunteer for. It is a life-safety device, its buttons are commonly covered or shrouded precisely because an accidental press causes an outage, and any work near an EPO circuit runs under the site's authorized procedure. In an interview, awareness of what it is and a clear statement that you would never operate or test one outside that procedure is the whole expected answer.
Sequence of operations SOO: The written description of how a system is supposed to behave: what starts, in what order, at what setpoint, on what signal, and what happens when something fails. It is the reference a functional test is written against.; The distinction that matters The sequence of operations says what should happen; the BMS is the machinery that makes it happen. A functional test that passes against a setpoint the sequence never specified proves nothing, which is why acceptance criteria have to trace back to the sequence rather than to whatever the controller currently holds.
Setpoint: The target value a control system is holding to, such as a supply air temperature or a pressure. The control system acts to keep the measured value at the setpoint.; The distinction that matters A setpoint is what the equipment is currently told to do. Design intent is what the building was meant to do. A test passed against the wrong setpoint is worse than a failed test, because it is recorded as a pass and the mismatch stays hidden until load arrives.
Permit to work: A formal, time-bound authorization to carry out a defined piece of work under stated controls, issued by someone with the authority to issue it and closed when the work is finished or the window ends.; The distinction that matters A permit authorizes; a method statement or MOP describes; a risk assessment justifies. Having done a job before is not authorization to do it here, and the sentence that reads well in an interview is that you would not start until the permit is issued and you have read what it actually covers.; Pick the governing permit in the free drill

Construction & site readiness

The build, fit-out, and handover language you'll meet on a data center project or contractor interview.

Site walk: A controlled walk through an area to observe progress, safety, access, and open items. In interviews the goal isn't to act like an inspector: it's to show what you notice, why it matters, who owns it, and what you must not touch.
CSA Civil, Structural & Architectural: The building-fabric side of a data center project (foundations, floors, walls, doors, and finishes) as opposed to the MEP (mechanical, electrical, plumbing) services that fill it.; Appears in: Contractor / CSA role guide · taught in Fundamentals
Punch list: The running list of incomplete or defective items found before handover, also called a snag list. Each item is logged, owned, corrected, and re-checked so a space can be signed off as complete.; Tag site findings in the free tag drill; Appears in: Trades on Site role guide · taught in Fundamentals
Temporary works: Engineered supports and structures (propping, hoarding, access platforms) that exist only during construction to keep work safe, then get removed before handover.; Pick the governing permit in the free drill; Appears in: Trades on Site role guide · taught in Fundamentals
MEWP Mobile Elevating Work Platform: A powered platform such as a scissor lift or boom used to work safely at height. Operating one needs training and authorization, and the area below is usually controlled.; Pick the governing permit in the free drill
Exclusion zone: A barriered-off area you must not enter without authorization, used to keep people clear of a hazard such as overhead work, live testing, or a lifting operation.; Pick the governing permit in the free drill
Penetration: A hole through a wall, floor, or fire barrier for cables, pipes, or ducts. Once services pass through, the gap must be sealed: an unsealed penetration in a fire-rated wall is a common observation.
Firestopping: The sealing of penetrations with rated materials so a fire barrier keeps its integrity. Missing or damaged firestopping is a quality and life-safety issue raised through the punch process.; Appears in: Trades on Site role guide · Contractor / CSA role guide · taught in Fundamentals
Commissioning Cx: The staged testing that proves systems work as designed before they go live, from individual components through to integrated, full-load scenarios. Data center practice normally runs the ladder from L0 design review to L6 handover; many people still say L1 to L5, which is the visible testing in the middle.; Walk L0 to L6 in the free ladder explorer; Appears in: QA/QC Quality Engineer role guide · Commissioning Engineer role guide · Controls / BMS Engineer role guide · taught in Fundamentals
Handover evidence: The documented proof that work is complete and compliant (test records, closed punch items, as-built drawings, and certificates) that transfers a space from the project team to operations.; Appears in: Contractor / CSA role guide · QA/QC Quality Engineer role guide · Commissioning Engineer role guide · taught in Fundamentals
ITP Inspection and Test Plan: The agreed record of what will be inspected, when, against what standard, and what evidence proves it: the contract between what a project builds and what it proves.; Set the point types in the free ITP builder; Appears in: Contractor / CSA role guide · QA/QC Quality Engineer role guide · taught in Fundamentals
Hold point: A stage in an ITP where work must stop and may not continue until an inspection is done and released, typically just before work becomes inaccessible or irreversible.; Set the point types in the free ITP builder; Appears in: Trades on Site role guide · QA/QC Quality Engineer role guide · taught in Fundamentals
NCR Non-Conformance Report: The document raised when work does not match its requirement, tracking a defect from raise through disposition, corrective action, and close, against the work rather than the person.; Tag site findings in the free tag drill; Appears in: Contractor / CSA role guide · QA/QC Quality Engineer role guide · taught in Fundamentals
Document number and revision: The coded identifier on a drawing or document and the tag that carries its current state, together telling you what a document is and whether it is the latest issued version to build from.
RFSU Ready For Start-Up: The declaration that a system is complete and safe to energize so start-up can begin. It closes the pre-functional level of the commissioning ladder and is expected to rest on closed checklists rather than on a target date.; The distinction that matters RFSU is an evidence statement, not a milestone someone hopes to hit. Declaring it early is one of the most expensive mistakes on a project, because start-up then runs on an unfinished installation and the defects reappear as functional test failures later, with more people waiting.; Walk L0 to L6 in the free ladder explorer
IST Integrated Systems Test: The test where all systems run together under failure scenarios, commonly including a black-building style test in which the facility loses utility power and must carry the load on its own systems, in sequence, without dropping it.; The distinction that matters Functional testing proves each system alone against its sequence. IST proves them together, and the interactions between systems are where facilities actually fail. That is why an IST still matters after every individual system has already passed.; Walk L0 to L6 in the free ladder explorer
Single-line diagram One-line: A simplified drawing that shows the electrical path from the utility intake through switchgear, generators, UPS and distribution down toward the load, using one line to represent a whole circuit.; The distinction that matters A single-line shows the path and how it can be reconfigured, not the physical layout or the cable routes. Being able to trace one source to one load, name what is between them, and say what happens when one element is out is a standard opening exercise in facilities and commissioning screens.; Trace it in the free power train explorer
MEP Mechanical, Electrical & Plumbing: The building services side of a project: cooling and mechanical plant, electrical systems and distribution, and the piped services that support them. The counterpart to CSA, which is the building fabric.
Witnessed result: A test result that an independent party watched and signed, rather than one reported by the party that performed the work.; The distinction that matters Self-reported passes are how failures hide, which is the whole reason witness points exist in an ITP. If you are asked what makes a record trustworthy, the answer is who watched it and what they signed, not how neat the paperwork looks.

Next · practice

Learn to read the environment, not just the definitions

Interviews test how you think when something is visible but not explained: a blocked route, an alarm, an unsealed penetration. The Question Bank drills the discipline behind real scenarios (observe, classify, escalate, document) with answer frameworks and red flags. Start with 25 free questions.

Get 25 Free Questions Find your path on Start Here