Skip to content

EN 50600 glossary

The EN 50600 series works with a fixed vocabulary that every part takes over from Part 1 and supplements with its own terms. This glossary explains the terms used on the part pages in our own words, with the reference in the standard; it does not replace the terms and definitions clauses of the individual parts.

  1. Part 1 specifies four Availability Classes, which are selected per infrastructure: for power supply and distribution, environmental control, and telecommunications cabling. Class 1 is a single path, Class 2 a single path with redundancy, Class 3 a multiple-path solution that allows repair during operation, and Class 4 is fault tolerant except during maintenance. A data center only achieves a class when every infrastructure meets or exceeds it. Percentage values are deliberately not part of the classification, because a power outage of a few seconds has different consequences than several minutes without cooling. The classes are comparable to the levels of other classifications, such as the Tier levels, but not identical to them; the standard describes outcomes, not individual criteria.

    Defined in: EN 50600‑1:2019, Clauses 5.1 and 7.2.2, Annex A.

  1. All KPIs of the EN 50600‑4 series treat the data center as a system with a boundary through which energy and water flow; the boundary is drawn around the data center at the energy supplier’s point of connection and is the same for PUE, REF, ERF, CUE, and WUE. All energy carriers that cross this boundary inward count toward the total energy consumption, including gas, heating oil, or delivered chilled water. In the sense of the ERF, the point of connection is the point on the boundary at which energy is measured and handed over to another party that uses it outside, such as a heating network. Reports on REF and ERF must state the boundaries considered so that the value can be traced.

    Defined in: EN 50600‑4‑6:2020, Clauses 3.1.2, 4 and 8.1.1; EN 50600‑4‑3:2016+A1:2019, Clause 8.

  2. Part 1 makes the business risk analysis a prerequisite for conformance. It combines two assessments: the business impact analysis, that is, the cost of downtime per event, and the risk analysis, in which risk is evaluated in a matrix (risk map) as the product of impact and probability. External threats to the site are explicitly included. The standard does not prescribe a method and refers to general standards on risk management. The results are the Availability Class per infrastructure, the Protection Classes per space, and the granularity level. Without a documented analysis, no class can be justified.

    Defined in: EN 50600‑1:2019, Clauses 4, 5.1 to 5.3 and 9.1.

  1. Part 2‑4 organizes the cabling into areas in accordance with EN 50173‑5. The entrance room (ER) houses the equipment of external telecommunications carriers and forms the demarcation points. The main distribution area (MDA) houses the main distributor with the central routers and switches and is the central distribution point. The intermediate distribution area (IDA) is optional and serves expansion or segmentation in large data centers. The zone distribution area (ZDA) houses the zone distributors, and the equipment distribution area (EDA) houses the equipment outlets, servers, and storage. The entrance room, the main distribution area, and the equipment distribution area must be present; the others are optional. The building entrance facility (BEF) is the point where cables enter the building.

    Defined in: EN 50600‑2‑4:2023, Clauses 3.1.2, 3.1.7, 3.1.9, 3.1.19, 3.1.24, 3.1.34 and 6.3.2.

  2. The carbon usage effectiveness (CUE) relates the CO2 emissions attributable to the data center in one year to the annual energy consumption of the IT equipment. The ideal value is zero; there is no upper limit. The standard distinguishes categories by the emission sources considered: category 1 considers only electrical energy from external and internal sources and only CO2, while category 2 additionally covers all other energy supplies and emission sources as CO2 equivalents, such as refrigerant losses. Derived KPIs, such as the partial CUE for subsystems or the design CUE for the planning phase, must be labeled as such.

    Defined in: EN 50600‑4‑8:2022, Clauses 5, 6.2.2 and 8.4. International equivalent ISO/IEC 30134‑8.

  3. The cooling efficiency ratio (CER) divides the annual quantity of heat removed by the annual energy consumption of the cooling systems, both measured in the same unit and for the same period. It is an efficiency in the sense of Part 4‑1, because input and output have the same unit. The standard describes derived quantities: the interim CER, the cooling performance ratio (CPR) as an instantaneous value calculated from heat load and electrical power, and the reciprocal CEF, which expresses the effort as a proportion of the heat removed and can be applied to subsystems of the cooling chain. Outdoor air temperature and part load have a strong influence on the values.

    Defined in: EN 50600‑4‑7:2020, Clauses 3.1.1 and 5.1, Annex D. International equivalent ISO/IEC 30134‑7.

  4. The cooling performance ratio (CPR) is the ratio of the actual heat load of the data center to the actual electrical power of the cooling systems, both in kilowatts and measured at the same point in time. Unlike the CER, which is an annual value, the CPR shows the state within a short time window and is suitable for capacity management and for plotting against outdoor air temperature; a CPR report must state the time and the conditions. The cooling efficiency factor (CEF) is the reciprocal of the CER and expresses what proportion of the heat removed the cooling system uses as electrical energy. As a partial CEF, it can be applied to any subsystem of the cooling chain, such as the computer room air conditioning unit, the chiller, and the dry cooler.

    Defined in: EN 50600‑4‑7:2020, Clauses 3.1.2 and 4, Annexes D.2 and D.3.

  1. Part 1 names the spaces a data center may contain and assigns functions to them: the computer room space for the IT equipment, the electrical space for switchgear, batteries, and UPS, the mechanical space for cooling and ventilation equipment, and the control room space as the central location for control and monitoring, plus main distributor, testing, office, storage, and holding spaces. Some spaces may be shared between the data center and the rest of the building, such as building entrance facilities and generator or transformer spaces. Whether spaces are separated depends on availability, security, and environmental conditions; a small data center may combine the computer room space and the electrical space in one room.

    Defined in: EN 50600‑1:2019, Clauses 3.1.7 to 3.1.35 and 6.2.

  2. The Technical Specification groups three KPIs under dependability. Reliability describes the ability to function as required throughout a time interval without failing; availability is the proportion of time during which the service is delivered at the operating point; the failure rate counts failures per unit of time. Each of them exists in three forms: past, measured over one year; inherent, calculated from the design under ideal conditions, for example availability from the mean time between failures (MTBF) and the mean time to recovery (MTTR); and operational, determined under real maintenance conditions from the mean time between maintenance (MTBM) and the mean down time (MDT). The German edition translates “reliability” as “Funktionsfähigkeit”, because “Zuverlässigkeit” is reserved for the umbrella term “dependability”.

    Defined in: CLC/TS 50600‑4‑31:2024 (Technical Specification), Clauses 3.1.9, 3.1.14, 3.1.22, 3.1.23, 3.1.28, 6.2.2, 6.2.4 and 6.3.

  3. A Committee Draft (CD) is a working draft of the working group within the technical committee; the national mirror committees comment on it, and there may be several versions (for example, a second CD). The enquiry draft prEN is the version that goes to all national committees as a draft European Standard and is open for public comment within a set period for objections; in Germany, it is published as E DIN EN. The final draft FprEN contains the incorporated comments and goes to the formal vote of the CENELEC members. None of these stages is a valid standard. For EN 50600‑2‑2 Edition 3, for example, the second CD was commented on in autumn 2025 and the prEN was submitted in January 2026.

    Defined in: CEN-CENELEC Internal Regulations; stages and deadlines on the CENELEC procedure pages (Enquiry, Formal Vote). Example: prEN 50600‑1:2026 (enquiry until 17 April 2026).

  1. The carbon dioxide emission factor (EFC) states the specific CO2 emissions resulting from the energy consumption of a data center and the operation of its facilities, expressed per kilowatt-hour; the operation of the facilities also includes emissions from refrigerants or diesel generators. The factors must be determined by the energy supplier and approved by the competent authority; where regional values are not available, national or international reference values apply, and the source must be stated in the CUE report. The carbon dioxide equivalent (CO2e) expresses the global warming potential of other greenhouse gases, such as refrigerants, in terms of the effect of one unit of carbon dioxide; CUE category 2 works with CO2e, category 1 only with CO2.

    Defined in: EN 50600‑4‑8:2022, Clauses 3.1.6, 3.1.7 and 6.2.2, Annex C.1. International equivalent ISO/IEC 30134‑8.

  2. Alongside availability and physical security, energy efficiency enablement is the third criterion by which Part 1 classifies a data center. It does not refer to a specific efficiency value but to the technical prerequisite for recording, calculating, and reporting energy consumption. Before planning starts, the owner determines the granularity level at which measurements are taken; the basis can be an operating cost analysis, operational processes according to Part 3‑1, KPIs of the 4-x series, or legal requirements. The design parts 2‑2, 2‑3, and 2‑4 derive the measurement points in power supply, cooling, and cabling from this.

    Defined in: EN 50600‑1:2019, Clauses 3.1.16 and 7.4.1.

  3. The energy reuse factor (ERF) relates the energy that the data center hands over annually to third parties outside its boundary, where it replaces other energy, to the annual total energy consumption. The most common example is waste heat that heats a neighboring building or drives an absorption chiller. What counts is the energy at the point of connection on the boundary, not the benefit to the recipient. Energy reused within the data center does not count, because it is already reflected in a lower PUE. The value range is between zero (no reuse) and one.

    Defined in: EN 50600‑4‑6:2020, Clauses 3.1.1, 4 and 5. International equivalent ISO/IEC 30134‑6.

  4. The energy water intensity factor (EWIF) describes how much water is consumed in generating the energy a data center obtains, expressed in cubic meters per megawatt-hour. It is calculated and provided by the energy supplier on the basis of its mix of power plants and cooling methods. For category 3 WUE, the total energy consumption of the data center is multiplied by the EWIF, so that water which does not physically arrive at the data center boundary but evaporates at the power plant also becomes visible. If the supplier does not provide an actual value, the calculation remains at category 1 or 2; the example values in the annex of the standard must not be used as a substitute in that case. The EWIF is therefore also a criterion for selecting the site and the energy mix.

    Defined in: EN 50600‑4‑9:2022, Clauses 3.1.9 and 6.2.2.4, Annex C.

  1. The granularity level specifies energy efficiency enablement in more detail. Level 1 measures the data center as a whole, Level 2 individual facilities, and Level 3 systems within spaces and supply paths. The owner determines the level before planning. Part 2‑2 translates it into measurement points in the power supply: for Level 1 at the supplies and at the output of the uninterruptible power system, for Level 2 at intermediate points of the distribution, and for Level 3 at the protected sockets of the tertiary distribution. The higher the level, the more precisely KPIs such as PUE can be determined and losses allocated to individual subsystems.

    Defined in: EN 50600‑1:2019, Clause 7.4.1; EN 50600‑2‑2:2019, Clause 8.1.

  1. Part 4‑1 sets out what a KPI in the EN 50600‑4 series must provide: a clearly defined resource usage, a formula, specified measurement points, an assessment period, and reporting requirements. The standard distinguishes effectiveness (input and output quantities have different units, for example liters per kilowatt-hour) and efficiency (same units). An important rule: the KPIs are not intended to be combined into an overall KPI or to be used to compare two data centers. The series also does not set any limit or target values.

    Defined in: EN 50600‑4‑1:2016, Clauses 3.1.2 to 3.1.4, 5.1, 5.4 and 5.5.

  1. Part 2‑5 requires physical security to be implemented according to a model of layered levels, which the standard calls the “onion skin” model. The Protection Class increases from the outside in; anyone with access to an area of a given class does not automatically have access to all areas of a lower class. Access remains limited to the operationally necessary minimum in terms of spaces, time, personnel, and knowledge. Because real buildings are rarely concentric, the standard allows the model to be represented as Protection Class islands. The model also applies to the Protection Classes against intrusion, and areas of the highest class must be surrounded in all directions by areas of the next lower class.

    Defined in: EN 50600‑2‑5:2021, Clauses 6.1.2, 7.1 and 9.2.4.

  2. The Technical Specification on KPIs for resilience relates every KPI to an operating point (OP). This can be a single socket, a server rack, or a row of racks at which power supply and environmental control must jointly deliver the agreed service; operation there only counts as successful if both infrastructures work successfully at the same time. The operating point is selected on the basis of the risk analysis, usually where the requirements for service quality are highest. Because the same infrastructure delivers different values at different operating points, every reported KPI must state the operating point and the assumed loads.

    Defined in: CLC/TS 50600‑4‑31:2024 (Technical Specification), Clauses 3.1.21, 4.2 and 6.2.2.

  1. A single point of failure (SPoF) is a functional element whose failure causes the overall system at the operating point to fail; a double point of failure (DPoF) is a pair of elements whose simultaneous failure has this effect. A single point of reduced availability (SPoRA) is an element whose failure does not cause the system to fail but violates the required operational availability; the DPoRA is the counterpart for double failures. All four are determined as integers from the resilience model; a lower value means better fault tolerance or availability tolerance. A failure can be an unplanned disruption or a planned maintenance event. According to the Technical Specification, the number of SPoRAs and DPoRAs says more about resilience than counting SPoFs alone.

    Defined in: CLC/TS 50600‑4‑31:2024 (Technical Specification), Clauses 3.1.31, 3.1.32, 5.5, 5.6, 6.4 and 6.5.

  2. The power usage effectiveness (PUE) divides the annual total energy consumption at the data center boundary, including all forms of energy such as electricity, gas, or delivered chilled water, by the annual energy consumption of the IT equipment. By definition, the value is greater than one. The standard specifies three measurement categories, which differ in where the IT energy is measured: at the UPS output, at the output of the power distribution unit, or at the input of the IT equipment. Derived KPIs are the partial PUE (pPUE) for a subsystem, the interim PUE (iPUE) for periods of less than one year, and the design PUE (dPUE) as a planning target before operation starts. Every reported value states its category and measurement period.

    Defined in: EN 50600‑4‑2:2016+A1:2019, Clauses 3.1.3 to 3.1.7, 5.1 and 6.2.1, Annex C. International equivalent ISO/IEC 30134‑2 (second edition 2026).

  3. Every space and every supply path is assigned a Protection Class, regardless of the size and purpose of the data center; a data center as a whole does not have one. The class is selected per type of event: unauthorized access, intrusion, and internal and external environmental events. For access, Part 2‑5 describes the classes as ascending levels of access control, from public areas to areas reserved for persons with a demonstrated need, with an escort requirement for everyone else. For internal events, the levels range from no requirement through detection to detection and mitigation within the space. The related construction requirements are set out in Part 2‑1.

    Defined in: EN 50600‑1:2019, Clauses 7.3.1 to 7.3.4; EN 50600‑2‑5:2021, Clauses 6.1.2, 7.1 and 9.1.

  1. If the formal vote on the final draft is positive, the text is ratified as a European Standard; the title page states the date of approval by CENELEC. The members are then obliged to give the standard the status of a national standard without any alteration. The European foreword sets two dates for this: the latest date of implementation at national level (dop) and the latest date by which conflicting national standards must be withdrawn (dow). In Germany, the standard is published as DIN EN 50600-x with a VDE number, with a national foreword and, where applicable, national footnotes by GK 719. Only this edition is the binding version.

    Defined in: EN 50600‑1:2019, title page of the EN and European foreword; EN 50600‑2‑1:2021, national foreword of the German edition.

  2. The standard does not describe redundancy using formulas such as N+1 or 2N but through the outcome that an Availability Class requires: whether a single fault in a functional element may lead to loss of function, whether planned maintenance must be possible without shutting down the load, and whether multiple paths are present. Class 2 of the power distribution is a single-path solution with redundant devices, Class 3 a multiple-path solution that allows repair during operation, and Class 4 is fault tolerant. Common market designations such as N+1 or 2N describe designs that can meet a class; the classes themselves are defined by the required behavior, not by a redundancy formula, and conformity is demonstrated via the class.

    Defined in: EN 50600‑1:2019, Clause 7.2.2; EN 50600‑2‑2:2019, Clauses 6.2.6 and 6.3.4.

  3. For the REF, only renewable energy counts for which the data center holds the valid rights to the environmental attributes of the renewable generation. For this purpose, the standard names the renewable energy certificate, a tradable contractual instrument that serves as evidence that a specific quantity of energy was generated from a renewable source. What counts is energy generated on site whose certificates were canceled there, purchased and consumed certificates, and the share of renewable energy stated in writing by the supplier; energy generated on site but sold on together with its certificates does not count. Commission Delegated Regulation (EU) 2024/1364 uses the term guarantee of origin for this and requires, for the European database on data centers, the renewable energy to be broken down into guarantees of origin, power purchase agreements, and on-site generation.

    Defined in: EN 50600‑4‑3:2016+A1:2019, Clauses 3.1.2, 3.1.3 and 5.1; Commission Delegated Regulation (EU) 2024/1364, Annex II.

  4. The renewable energy factor (REF) relates the renewable energy owned and controlled by a data center to its total annual energy consumption. Only energy for which the rights to the environmental attributes lie with the data center is counted: energy generated on site whose certificates were canceled there, purchased and consumed certificates, and the share of renewable energy stated in writing by the supplier. Energy generated on site but sold on does not count. The REF can therefore reach at most 1.00. The KPI serves to monitor the use of renewable energy and to reduce dependence on a single energy source.

    Defined in: EN 50600‑4‑3:2016+A1:2019, Clauses 3.1.3, 4 and 5.1. International equivalent ISO/IEC 30134‑3.

  5. According to the Technical Specification, resilience is the capacity to withstand disruptive events or to limit their extent or duration, which includes being able to anticipate them, absorb them, and recover from them. A resilience level (RL) lists the attributes of a defined service at an operating point: the reporting interval in years, the maximum accepted number of service violations, and the maximum accepted down time per event, plus the permissible number of single points of failure; the operational availability follows from these. The normal resilience level (NRL) applies when all functional elements are in the up state; for periods of planned maintenance, a reduced resilience level (RRL) with lower requirements should be defined. If a service level agreement exists, its details are included in the definition.

    Defined in: CLC/TS 50600‑4‑31:2024 (Technical Specification), Clauses 3.1.24 to 3.1.26, 5.6.1 and 6.6.

  1. CLC/TC 215 “Electrotechnical aspects of telecommunication equipment” is the CENELEC technical committee responsible for the series; the texts are drafted in its working group WG 3 “Facilities and infrastructures”, and DKE holds the secretariat. DKE/GK 719 “Rechenzentren” (data centers) is the German mirror committee within DIN and VDE: it prepares the German comments and adopts the parts as DIN EN 50600 (VDE 0801‑600). Its six working groups GAK 719.0.1 to 719.0.6 are organized by discipline, from building construction to resilience KPIs. ISO/IEC JTC 1/SC 39 “Sustainability, IT and data centres” carries the content forward internationally as ISO/IEC 22237.

    Defined in: EN 50600‑1:2019, European foreword and national foreword of the German edition; ISO/IEC 22237‑1:2021, Foreword; DKE committee page GK 719 and CENELEC committee page CLC/TC 215 (accessed 17 September 2026).

  1. A Technical Report (CLC/TR) contains recommended practices and guidance, not requirements; in the EN 50600 series, these are the 99-x parts, such as the application guide 99‑3. A Technical Specification (CLC/TS) is the result of standardization work that is not yet published as a standard, either because of reservations about its content or because a different procedure was followed; in Germany, it is published as a VDE pre-standard (Vornorm), which is not part of the German body of standards. Members must make it available but may retain conflicting national standards. Examples are CLC/TS 50600‑2‑10, 50600-4‑31, and 50600-5‑1.

    Defined in: EN 50600‑1:2019, Introduction; CLC/TS 50600‑4‑31:2024, national foreword of the German edition and title page of the CLC/TS.

  1. The water reuse factor (WRF) relates the water a data center reuses annually to the water supplied to it annually. Reused water is water that leaves the data center boundary and serves a use unrelated to the data center; the standard distinguishes industrial reuse, for example for manufacturing, cleaning, or cooling, and non-industrial reuse. The value range is between zero and one; the ideal value of one means that all water used for operation is reused. The WRF complements the WUE, in whose categories 2 and 3 reuse reduces the water usage.

    Defined in: EN 50600‑4‑9:2022, Clauses 3.1.6 to 3.1.8 and 8.5.

  2. Water availability is the amount of renewable freshwater available per person and year at a location; water stress describes whether the freshwater needs of people and the environment can be met from it. The standard requires water availability to be reported as the Falkenmark water stress indicator, that is, as annual surface runoff relative to the population, and divides it into levels from water stress to absolute water scarcity. For category 3 WUE, the indicator belongs in the report as supplementary information, together with the land use of the data center, because the same water consumption must be assessed differently at a site where water is scarce than at a site where it is plentiful. The indicator is therefore one of the criteria the standard names for site selection.

    Defined in: EN 50600‑4‑9:2022, Clauses 3.1.10, 3.1.11 and 7, Annex B.

  3. The water usage effectiveness (WUE) divides the annual water usage of the data center by the annual energy consumption of the IT equipment. Water usage is the difference between water input and water output at the data center boundary. A value of zero means that no water is consumed for operation; there is no upper limit. As with the PUE, there are derived KPIs: interim, partial for subsystems, as a design value, and quality-related. The evidence that must be available on request also includes the associated PUE value with its category; for the design WUE, dWUE and dPUE should be set out in a common plan.

    Defined in: EN 50600‑4‑9:2022, Clauses 3.1.3, 5, 7, 8.1 and 8.4. International equivalent ISO/IEC 30134‑9.

  • DIN EN 50600‑1:2019‑08 (German edition of EN 50600‑1:2019), Clauses 3.1, 4, 5, 6.2, 7.2 to 7.4, 9.1, Annex A, Introduction, European and national foreword, title page
  • DIN EN 50600‑2‑1:2021‑09 (German edition of EN 50600‑2‑1:2021), national foreword
  • DIN EN 50600‑2‑2:2019‑08 (German edition of EN 50600‑2‑2:2019), Clauses 6.2.6, 6.3.4, 8.1
  • DIN EN 50600‑2‑4:2023‑09 (German edition of EN 50600‑2‑4:2023), Clauses 3.1, 6.3.2
  • DIN EN 50600‑2‑5:2021‑09 (German edition of EN 50600‑2‑5:2021), Clauses 6.1.2, 7.1, 9.1, 9.2.4
  • DIN EN 50600‑4‑1:2017‑06 (German edition of EN 50600‑4‑1:2016), Clauses 3.1, 5.1, 5.4, 5.5
  • DIN EN 50600‑4‑2:2019‑08 (German edition of EN 50600‑4‑2:2016 with corrigendum AC:2017 and amendment A1:2019), Clauses 3.1, 5.1, 6.2.1, Annex C
  • DIN EN 50600‑4‑3:2019‑08 (German edition of EN 50600‑4‑3:2016 with amendment A1:2019), Clauses 3.1, 4, 5.1, 8
  • DIN EN 50600‑4‑6:2020‑07 (German edition of EN 50600‑4‑6:2020), Clauses 3.1, 4, 5, 8.1.1
  • DIN EN 50600‑4‑7:2020‑11 (German edition of EN 50600‑4‑7:2020), Clauses 3.1, 4, 5.1, Annex D
  • DIN EN 50600‑4‑8:2023‑08 (German edition of EN 50600‑4‑8:2022), Clauses 3.1, 5, 6.2.2, 8.4, Annex C.1
  • DIN EN 50600‑4‑9:2023‑08 (German edition of EN 50600‑4‑9:2022), Clauses 3.1, 5, 6.2.2.4, 7, 8.1, 8.4, 8.5, Annexes B and C
  • DIN CLC/TS 50600‑4‑31:2025‑02 (German edition of CLC/TS 50600‑4‑31:2024, VDE pre-standard), national foreword and title page, Clauses 3.1, 4.2, 5.5, 5.6, 6.2 to 6.6
  • Commission Delegated Regulation (EU) 2024/1364 of 14 March 2024, Annex II
  • ISO/IEC 22237‑1:2021, Foreword
  • DKE committee page GK 719, CENELEC committee page CLC/TC 215, CENELEC BOSS (Enquiry, Formal Vote), all accessed 17 September 2026

Last reviewed on  · DCE academy

Contact

Questions about applying EN 50600?

Thomas Wawra answers your questions about applying EN 50600 in your project, and which seminar suits it, by phone or email.

Newsletter

Expert articles and standards updates by email

Expert articles from the newsroom, news from standardization, and the next seminar dates.

You will receive a confirmation email and can unsubscribe at any time. Privacy information