EPI Data Centre Associate key terms

A four-level rating scheme describing how much redundancy and fault tolerance a data centre's power, cooling and distribution paths provide, from a basic single-path design up to fully fault-tolerant concurrent maintainability.

The ability to service, repair or replace any component of the critical power or cooling path without interrupting the IT load, achieved through redundant capacity and isolation points.

Provision of extra capacity beyond the minimum needed to carry the load. N is the requirement, N+1 adds one spare unit, and 2N provides two complete independent systems.

The property of a system that lets it keep serving the IT load after one or more component failures, usually by switching automatically to a redundant path or unit.

Any component whose failure would interrupt the IT load because no alternative path or spare exists, such as a shared switchboard, pump or upstream transformer.

A battery-backed system placed between the utility supply and IT equipment that supplies power during outages and conditions electricity to protect against sags, surges and transients.

A solid-state device that transfers a load between two independent power sources within milliseconds, typically used to move equipment between UPS outputs without a break.

A switch that detects loss of the preferred power source and automatically transfers the load to an alternate source, commonly utility to generator, with a mechanical break.

An engine-driven machine that produces electrical power on site when the utility supply fails, sized to carry critical and often some non-critical loads for a defined runtime.

The ratio of total facility energy to IT equipment energy over a period. A value of 1.0 means all energy reaches IT; higher values indicate overhead from cooling, losses and other systems.

A layout and physical barrier strategy that separates cool supply air from hot exhaust air in the server room, improving cooling efficiency and reducing bypass and recirculation.

Software modelling of air or liquid flow, temperature and pressure in a data hall, used to predict hot spots, test layouts and validate cooling capacity before changes are made.

The practice of controlling where cool air goes and where hot air returns, using blanking panels, seals, raised-floor tiles, containment and fan control to avoid mixing and bypass.

Recommended ranges for temperature and humidity at IT equipment air intakes, published by industry bodies, that balance equipment reliability, energy use and operator comfort.

The ratio of water consumed by the facility, usually for evaporative cooling, to IT equipment energy over a period, expressed in litres per kilowatt-hour.

A planned, standards-based cabling system with defined hierarchy, media, connectors and documentation that supports current and future network and management connections in the facility.

The physical routing, support, labelling and separation of copper and fibre cabling to control airflow blockage, reduce strain, prevent interference and enable tracing and maintenance.

A point where cabling from different areas or carriers is terminated and patched together, such as a meet-me-room or main distribution area, enabling flexible interconnection.

The physical and logical links, including cross-connects and peering, that connect a facility's customers to carriers, cloud providers and other data centres.

A widely used industry standard covering data centre infrastructure, including site topology, cabling, redundancy and environmental requirements, with its own reliability rating levels.

A European series of standards for data centre facilities and infrastructure that defines availability classes, protection classes, physical security and energy efficiency management.

An international standard series for data centre facilities and infrastructure that aligns closely with EN 50600, covering design, availability, security and energy management.

A graded category in EN 50600 and ISO/IEC 22237 describing how resilient a facility is, ranging from a basic single-path design to a fault-tolerant, concurrently maintainable one.

An international standard for establishing, operating and continually improving an information security management system, applied to data centre services through certification audits.

An independent attestation report on a service organisation's controls over security, availability, processing integrity, confidentiality and privacy, widely requested by data centre customers.

An international standard for energy management systems that helps a facility set energy policy, measure consumption, identify savings and demonstrate continual improvement.

A systematic, documented examination of records, controls and practices against a standard, regulation or internal policy, producing findings and corrective actions.

A contract defining measurable service commitments such as availability, power and cooling delivery, response times and remedies when targets are missed.

An internal or contractual target for a specific service metric, such as 99.99% power availability, that supports and is usually stricter than the customer-facing agreement.

A controlled process for planning, reviewing, approving, implementing and reviewing changes to infrastructure or configuration, reducing unintended outages and preserving records.

A step-by-step written instruction for performing a specific task safely, including prerequisites, sequence, verification points, rollback steps and required approvals.

A documented routine instruction describing how to carry out a recurring operational activity consistently, such as a generator test or a customer access request.

Scheduled servicing of equipment and systems, such as UPS battery checks and chiller servicing, performed at defined intervals to prevent failures and keep warranties valid.

The average time taken to restore a failed component or service to normal operation, including detection, response, diagnosis and repair, used to measure maintainability.

The process of detecting, logging, escalating, resolving and reviewing unplanned events or outages, with the aim of restoring service quickly and preventing recurrence.

The discipline of finding and eliminating the underlying causes of recurring incidents, using trend analysis, root cause investigation and permanent corrective actions.

A structured investigation, often using techniques such as five whys or fishbone diagrams, to identify the fundamental reason an incident occurred and prevent repetition.

A repository recording the components of the environment and their relationships, used to understand dependencies, assess change impact and support incident diagnosis.

A maintained inventory of physical and logical assets with identifiers, location, ownership, lifecycle dates and warranty details, supporting maintenance, audit and replacement planning.

A scheduled process of filtering, testing and treating stored generator fuel to remove water, sediment and microbial growth, keeping it within specification for reliable engine operation.

A documented evaluation identifying hazards in a task, judging likelihood and severity, and recording controls and safe working steps before work starts.

An assessment of critical business processes and the effect of their disruption over time, used to set recovery priorities and justify resilience investment.

The maximum acceptable time to restore a process or service after disruption, driving design of redundancy, spares and recovery procedures.

The maximum acceptable amount of data loss measured in time before a disruption, which determines backup frequency and replication design.

A management system for building and testing an organisation's ability to keep critical services running during and after disruption, covering plans, roles, exercises and review.

A planned exercise that validates recovery plans and technical failover, ranging from tabletop walkthroughs to full live switchovers, with documented results and lessons learned.

A structured list of legal, regulatory, contractual and standards obligations applicable to the facility, with owners, evidence, review dates and status of fulfilment.

A defined boundary, such as a fence, wall or locked door, that separates areas of different access rights and is protected by controls including access control and monitoring.

The systems and rules that decide who may enter which areas and when, using credentials, biometrics, mantrap doors and audit logs to enforce authorised access only.

Camera monitoring of entrances, halls and critical plant, recorded and retained for investigation and deterrence, with privacy and retention rules applied.

The proportion of time a service is usable, often expressed as a percentage such as 99.99%, calculated from uptime against total agreed service time.