Help Print this page 

Document 32014D0202

Title and reference
Commission Decision of 20 March 2014 determining the European Union position for a decision of the Management entities under the Agreement between the Government of the United States of America and the European Union on the coordination of energy-efficiency labelling programmes for office equipment on adding specifications for computer servers and uninterruptible power supplies to Annex C to the Agreement and on the revision of specifications for displays and imaging equipment included in Annex C to the Agreement (Text with EEA relevance) (2014/202/EU)
  • In force
OJ L 114, 16.4.2014, p. 68–148 (BG, ES, CS, DA, DE, ET, EL, EN, FR, HR, IT, LV, LT, HU, MT, NL, PL, PT, RO, SK, SL, FI, SV)

ELI: http://data.europa.eu/eli/dec/2014/202/oj
Languages, formats and link to OJ
BG ES CS DA DE ET EL EN FR GA HR IT LV LT HU MT NL PL PT RO SK SL FI SV
HTML html BG html ES html CS html DA html DE html ET html EL html EN html FR html HR html IT html LV html LT html HU html MT html NL html PL html PT html RO html SK html SL html FI html SV
PDF pdf BG pdf ES pdf CS pdf DA pdf DE pdf ET pdf EL pdf EN pdf FR pdf HR pdf IT pdf LV pdf LT pdf HU pdf MT pdf NL pdf PL pdf PT pdf RO pdf SK pdf SL pdf FI pdf SV
Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal Display Official Journal
 To see if this document has been published in an e-OJ with legal value, click on the icon above (For OJs published before 1st July 2013, only the paper version has legal value).
Multilingual display
Text

16.4.2014   

EN

Official Journal of the European Union

L 114/68


COMMISSION DECISION

of 20 March 2014

determining the European Union position for a decision of the Management entities under the Agreement between the Government of the United States of America and the European Union on the coordination of energy-efficiency labelling programmes for office equipment on adding specifications for computer servers and uninterruptible power supplies to Annex C to the Agreement and on the revision of specifications for displays and imaging equipment included in Annex C to the Agreement

(Text with EEA relevance)

(2014/202/EU)

THE EUROPEAN COMMISSION,

Having regard to the Treaty on the Functioning of the European Union,

Having regard to Council Decision 2013/107/EU of 13 November 2012 on the signing and conclusion of the Agreement between the Government of the United States of America and the European Union on the coordination of energy-efficiency labelling programmes for office equipment (1), and in particular Article 4 thereof,

Whereas:

(1)

The Agreement provides for the European Commission, together with the United States Environmental Protection Agency (EPA), to develop and periodically revise common specifications for office equipment specification, thereby amending Annex C to the Agreement.

(2)

The position of the European Union with regard to amendment of the specifications is to be determined by the Commission.

(3)

The measures provided for in this Decision take account of the opinion given by the European Union Energy Star Board referred to in Article 8 of Regulation (EC) No 106/2008 of the European Parliament and of the Council of 15 January 2008 on a Community energy-efficiency labelling programme for office equipment (2) as amended by Regulation (EU) No 174/2013 (3).

(4)

The display specification in Annex C, Part II, and imaging equipment specification in Annex C, Part III, should be repealed and replaced by the specifications annexed to this Decision,

HAS ADOPTED THIS DECISION:

Sole Article

The position to be adopted by the European Union for a decision by the Management Entities under the Agreement between the Government of the United States of America and the European Union on the coordination of energy-efficiency labelling programmes for office equipment on revising the display and imaging equipment specifications in Annex C, Parts II and III, and adding new specifications for computer servers and uninterruptible power supplies, to the Agreement shall be based on the attached draft decision.

This Decision shall enter into force on the twentieth day following that of its publication in the Official Journal of the European Union.

Done at Brussels, 20 March 2014.

For the Commission

The President

José Manuel BARROSO


(1)  OJ L 63, 6.3.2013, p. 5.

(2)  OJ L 39, 13.2.2008, p. 1.

(3)  OJ L 63, 6.3.2013, p. 1.


ANNEX I

DRAFT DECISION

of …

of the Management entities under the Agreement between the Government of the United States of America and the European Union on the coordination of energy-efficiency labelling programmes for office equipment on adding specifications for computer servers and uninterruptible power supplies to Annex C to the Agreement and on the revision of specifications for displays and imaging equipment included in Annex C of the Agreement

THE MANAGEMENT ENTITIES,

Having regard to the Agreement between the Government of the United States and the European Union on the coordination of energy-efficiency labelling programmes for office equipment, and in particular Article XII thereof,

Whereas specifications for new products ‘computer servers’ and ‘uninterruptible power supplies’ should be added to the Agreement and existing specifications for product type ‘imaging equipment’ and ‘displays’ should be revised,

HAVE DECIDED AS FOLLOWS:

Part I ‘Displays’, Part II ‘Uninterruptible Power Supplies’, Part III ‘Computer Servers’ and Part IV ‘Imaging Equipment’ shall be added to Annex C of the Agreement between the Government of the United States and the European Union on the coordination of energy-efficiency labelling programmes for office equipment as laid down hereafter.

Part II ‘Displays’ and Part III ‘Imaging Equipment’ currently included in Annex C of the Agreement between the Government of the United States and the European Union on the coordination of energy-efficiency labelling programmes for office equipment shall be repealed.

The Decision shall enter into force on the twentieth day following its publication. The Decision, done in duplicate, shall be signed by the Co-chairs.

Signed in Washington DC on the […]

[…]

on behalf of the United States Environmental Protection Agency

Signed in Brussels on the […]

[…]

on behalf of the European Union


ANNEX II

ANNEX C

PART II TO THE AGREEMENT

'I.   DISPLAY SPECIFICATIONS

1.   Definitions

1.1.   Product Types

Electronic Display (Display): A commercially-available product with a display screen and associated electronics, often encased in a single housing, that as its primary function displays visual information from (1) a computer, workstation or server via one or more inputs (e.g., VGA, DVI, HDMI, Display Port, IEEE 1394, USB), (2) external storage (e.g., USB flash drive, memorycard), or (3) a network connection.

(a)

Computer Monitor: An electronic device, typically with a diagonal screen size greater than 12 inches and a pixel density greater than 5 000 pixels per square inch (pixels/in2), that displays a computer’s user interface and open programs, allowing the user to interact with the computer, typically using a keyboard and mouse.

Enhanced-Performance Display: A computer monitor that has all of the following features and functionalities:

(i)

A contrast ratio of at least 60:1 measured at a horizontal viewing angle of at least 85°, with or without a screen cover glass;

(ii)

A native resolution greater than or equal to 2,3 megapixels (MP); and,

(iii)

A colour gamut size of at least sRGB as defined by IEC 619662-1. Shifts in colour space are allowable as long as 99 % or more of defined sRGB colours are supported.

(b)

Digital Picture Frame: An electronic device, typically with a diagonal screen size less than 12 inches, whose primary function is to display digital images. It may also feature a programmable timer, occupancy sensor, audio, video, or bluetooth or wireless connectivity.

(c)

Signage Display: An electronic device typically with a diagonal screen size greater than 12 inches and a pixel density less than or equal to 5 000 pixels/in2. It is typically marketed as commercial signage for use in areas where it is intended to be viewed by multiple people in non-desk based environments, such as retail or department stores, restaurants, museums, hotels, outdoor venues, airports, conference rooms or classrooms.

1.2.   External Power Supply (EPS): Also referred to as an external power adapter. A component contained in a separate physical enclosure external to a display, designed to convert line voltage ac input from the mains to lesser dc voltage(s) in order to provide power to the display. An EPS connects to the display via a removable or hard-wired male/female electrical connection, cable, cord or other wiring.

1.3.   Operational Modes:

(a)

On Mode: The power mode in which the product has been activated, and is providing one or more of its principal functions. The common terms, ‘active,’‘in-use,’ and ‘normal operation’ also describe this mode. The power in this mode is typically greater than the power in Sleep Mode and Off Mode.

(b)

Sleep Mode: The power mode the product enters after receiving a signal from a connected device or an internal stimulus. The product may also enter this mode by virtue of a signal produced by user input. The product must wake on receiving a signal from a connected device, a network, a remote control, and/or an internal stimulus. While the product is in this mode, it is not producing a visible picture, with the possible exception of user-oriented or protective functions such as product information or status displays, or sensor-based functions.

Notes:

1.

Examples of internal stimuli are a timer or occupancy sensor.

2.

A power control is not an example of user input.

(c)

Off Mode: The power mode in which the product is connected to a power source, and is not providing any On Mode or Sleep Mode functions. This mode may persist for an indefinite time. The product may only exit this mode by direct user actuation of a power switch or control. Some products may not have this mode.

1.4.   Luminance: The photometric measure of the luminous intensity per unit area of light travelling in a given direction, expressed in candelas per square meter (cd/m2). Luminance refers to the brightness settings of a display.

(a)

Maximum Reported Luminance: The maximum luminance the display may attain at an On Mode preset setting, and as specified by the manufacturer, for example, in the user manual.

(b)

Maximum Measured Luminance: The maximum luminance the display may attain by manually configuring its controls, such as brightness and contrast.

(c)

As-shipped Luminance: The luminance of the display at the factory default preset setting the manufacturer selects for normal home or applicable market use. The As-shipped Luminance of displays with Automatic Brightness Control (ABC) enabled by default may vary based on the Ambient Light Conditions of the location in which the display is installed.

1.5.   Screen Area: The viewable screen width multiplied by the viewable screen height, expressed in square inches (in2).

1.6.   Automatic Brightness Control (ABC): The self-acting mechanism that controls the brightness of a display as a function of ambient light.

1.7.   Ambient Light Conditions: The combination of light illuminances in the environment surrounding a display, such as a living room or an office.

1.8.   Bridge Connection: A physical connection between two hub controllers, typically, but not limited to, USB or FireWire, which allows for expansion of ports typically for the purpose of relocating the ports to a more convenient location or increasing the number of available ports.

1.9.   Network capability: An ability to obtain an IP address when connected to a network.

1.10.   Occupancy Sensor: A device used to detect human presence in front of or in the area surrounding a display. An occupancy sensor is typically used to switch a display between On Mode and Sleep or Off Mode.

1.11.   Product Family: A group of displays, made under the same brand, sharing a screen of the same size and resolution, and encased in a single housing that may contain variations in hardware configurations.

Example: Two computer monitors from the same model line with a diagonal screen size of 21 inches and a resolution of 2,074 megapixels (MP), but with variations in features such as built-in speakers or camera, could be qualified as a product family.

1.12.   Representative Model: The product configuration that is tested for ENERGY STAR qualification and is intended to be marketed and labelled as ENERGY STAR.

2.   Scope

2.1.   Included Products

2.1.1.

Products that meet the definition of a display as specified herein and are powered directly from ac mains, via an external power supply, or via a data or network connection, are eligible for ENERGY STAR qualification, with the exception of products listed in Section 2.2.

2.1.2.

Typical products that would be eligible for qualification under this specification include:

(a)

Computer Monitors

(b)

Digital Picture Frames

(c)

Signage Displays, and,

(d)

Additional products including monitors with keyboard, video and mouse (KVM) switch functionality, and other industry-specific displays that meet the definitions and qualification criteria in this specification.

2.2.   Excluded Products

2.2.1.

Products that are covered under other ENERGY STAR product specifications are not eligible for qualification under this specification. The list of specifications currently in effect can be found at www.eu-energystar.org.

2.2.2.

The following products are not eligible for qualification under this specification:

(a)

Products with a viewable diagonal screen size greater than 61 inches;

(b)

Products with an integrated television tuner;

(c)

Products that are marketed and sold as televisions, including products with a computer input port (e.g., VGA) that are marketed and sold primarily as televisions;

(d)

Products that are component televisions. A component television is a product that is composed of two or more separate components (e.g., display device and tuner) that are marketed and sold as a television under a single model or system designation. A component television may have more than one power cord;

(e)

Dual-function televisions/computer monitors that are marketed and sold as such;

(f)

Mobile computing and communication devices (e.g., tablet computers, slates, electronic readers, smartphones);

(g)

Products that must meet specifications for medical devices that prohibit power management capabilities and/or do not have a power state meeting the definition of Sleep Mode; and,

(h)

Thin clients, ultra-thin clients, or zero clients.

3.   Qualification Criteria

3.1.   Significant Digits and Rounding

3.1.1.

All calculations shall be carried out with directly measured (unrounded) values.

3.1.2.

Unless otherwise specified, compliance with specification requirements shall be evaluated using directly measured or calculated values without any benefit from rounding.

3.1.3.

Directly measured or calculated values that are submitted for reporting on the ENERGY STAR website shall be rounded to the nearest significant digit as expressed in the corresponding specification requirements.

3.2.   General Requirements

3.2.1.

External Power Supply: If the product is shipped with an EPS, the EPS shall meet the level V performance requirements under the International Efficiency Marking Protocol, and include the level V marking. Additional information on the Marking Protocol is available at www.energystar.gov/powersupplies

External Power Supplies shall meet level V requirements when tested using the Test Method for Calculating the Energy Efficiency of Single-Voltage External Ac-Dc and Ac-Ac Power Supplies, Aug. 11, 2004

3.2.2.

Power Management:

(a)

Products shall offer at least one power management feature that is enabled by default, and that can be used to automatically transition from On Mode to Sleep Mode either by a connected host device or internally (e.g., support for VESA Display Power Management Signalling (DPMS), enabled by default).

(b)

Products that generate content for display from one or more internal sources shall have a sensor or timer enabled by default to automatically engage Sleep or Off Mode.

(c)

For products that have an internal default delay time after which the product transitions from On Mode to Sleep Mode or Off Mode, the delay time shall be reported.

(d)

Computer monitors shall automatically enter Sleep Mode or Off Mode within 15 minutes of being disconnected from a host computer.

3.3.   On Mode Requirements

3.3.1.

On Mode power (PON), as measured per the ENERGY STAR test method shall be less than or equal to the Maximum On Mode Power Requirement (PON_MAX), as calculated and rounded per Table 1, below.

If the product’s pixel density (DP), as calculated per Equation 1, is greater than 20 000 pixels/in2, then the screen resolution (r) used to calculate PON_MAX shall be determined per Equation 2.

Equation 1: Calculation of Pixel Density

Formula

Where

DP is the pixel density of the product rounded to the nearest integer, in pixels/in2,

r is the screen resolution, in megapixels, and

A is the viewable screen area, in in2.

Equation 2: Calculation of Resolution if the Product’s Pixel Density (DP) Exceeds 20 000 pixels/in2

Formula

Formula

Where:

r1 and r2 are the screen resolutions, in megapixels, to be used when calculating PON_MAX,

DP is the pixel density of the product rounded to the nearest integer, in pixels/in2, and

A is the viewable screen area, in in2.

Table 1

Calculation of Maximum On Mode Power Requirements (PON_MAX)

Product Type and

Diagonal Screen Size, d

(in inches)

PON_MAX

where Dp ≤ 20 000 pixels/in2

(in watts)

Where:

r = Screen resolution in megapixels

A = Viewable screen area in in2

The result shall be rounded to the nearest tenth of a watt

PON_MAX

where DP > 20 000 pixels/in2

(in watts)

Where:

r = Screen resolution in megapixels

A = Viewable screen area in in2

The result shall be rounded to the nearest tenth of a watt

d < 12,0

Formula

Formula

12,0 ≤ d < 17,0

Formula

Formula

17,0 ≤ d < 23,0

Formula

Formula

23,0 ≤ d < 25,0

Formula

Formula

25,0 ≤ d ≤ 61,0

Formula

Formula

30,0 ≤ d ≤ 61,0

(for products meeting the definition of a Signage Display only)

Formula

Formula

3.3.2.

For products meeting the definition of an Enhanced-Performance Display, a power allowance (PEP), as calculated per Equation 3, shall be added to PON_MAX, as calculated per Table 1. In this case, PON, as measured per the ENERGY STAR test method shall be less than or equal to the sum of PON_MAX and PEP.

Equation 3: Calculation of On Mode Power Allowance for Enhanced-Performance Displays

Formula

Formula

Where:

PEP < 27″ is the On Mode power allowance, in watts, for an Enhanced-Performance Display with a diagonal screen size less than 27 inches,

PEP ≥ 27″ is the On Mode power allowance, in watts, for an Enhanced-Performance Display with a diagonal screen size greater than or equal to 27 inches, and

PON_MAX is the maximum On Mode power requirement, in watts.

3.3.3.

For products with Automatic Brightness Control (ABC) enabled by default, a power allowance (PABC), as calculated per Equation 5, shall be added to PON_MAX, as calculated per Table 1, if the On Mode power reduction (RABC), as calculated per Equation 4, is greater than or equal to 20 %.

(a)

If RABC is less than 20 %, PABC shall not be added to PON_MAX.

(b)

PON, as measured with ABC disabled per the ENERGY STAR test method shall be less than or equal to PON_MAX.

Equation 4: Calculation of On Mode Power Reduction for Products with ABC Enabled by Default

Formula

Where

RABC is the On Mode percent power reduction due to ABC,

P300 is the measured On Mode power, in watts, when tested with an ambient light level of 300 lux, and

P10 is the measured On Mode power, in watts, when tested with an ambient light level of 10 lux.

Equation 5: Calculation of On Mode Power Allowance for Products with ABC Enabled by Default

Formula

Where:

PABC is the On Mode power allowance, in watts, and

PON_MAX is the maximum On Mode power requirement, in watts.

3.3.4.

For products powered with a low-voltage dc source, PON, as calculated per Equation 6, shall be less than or equal to PON_MAX, as calculated per Table 1.

Equation 6: Calculation of On Mode Power for Products Powered by a Low-voltage Dc Source

Formula

Where:

PON is the calculated On Mode power, in watts,

PL is the ac power consumption, in watts, of the low-voltage dc source with the unit under test (UUT) as the load, and

PS is the marginal loss of the ac power supply of the source, in watts.

3.4.   Sleep Mode Requirements

3.4.1.

Measured Sleep Mode power (PSLEEP) for products with none of the data or network capabilities included in Table 3 or 4 shall be less than or equal to the Maximum Sleep Mode Power Requirement (PSLEEP_MAX), as specified in Table 2.

Table 2

Maximum Sleep Mode Power Requirement (PSLEEP_MAX)

PSLEEP_MAX

(watts)

0,5

3.4.2.

Measured Sleep Mode power (PSLEEP) for products with one or more of the data or network capabilities included in Table 3 or 4 shall be less than or equal to the Maximum Data/Networking Sleep Mode Power Requirement (PSLEEP_AP), as calculated per Equation 7.

Equation 7: Calculation of Maximum Data/Networking Sleep Mode Power requirement

Formula

Where:

PSLEEP_AP is the Maximum Sleep Mode Power Requirement, in watts, for products that were tested with additional power-consuming capabilities,

PSLEEP_MAX is the Maximum Sleep Mode Power Requirement, in watts, as specified in Table 2,

PDN is the power allowance, in watts, as specified in Table 3 for data or networking capability connected during Sleep Mode testing, and

PADD is the power allowance, in watts, as specified in Table 4 for additional capabilities enabled by default that are active during Sleep Mode testing.

Table 3

Power Allowances in Sleep Mode for Data or Network Capabilities

Capability

Included Types

PDN

(watts)

 

USB 1.x

0,1

USB 2.x

0,5

USB 3.x, DisplayPort (non-video connection), Thunderbolt

0,7

Network

Fast Ethernet

0,2

Gigabit Ethernet

1,0

Wi-Fi

2,0


Table 4

Power Allowances in Sleep Mode for Additional Capabilities

Capability

Included Types

PADD

(watts)

Sensor

Occupancy Sensor

0,5

Memory

Flash memory-card/smart-card readers, camera interfaces, PictBridge

0,2

Example 1: A digital picture frame with only one bridging or network capability connected and enabled during Sleep Mode testing, Wi-Fi, and no additional capabilities enabled during Sleep Mode testing, would qualify for the 2,0 W Wi-Fi adder. Recalling that Formula, Formula.

Example 2: A computer monitor with USB 3.x and DisplayPort (non-video connection) bridging capability shall be tested with only the USB 3.x connected and enabled. Assuming no additional capabilities are enabled during Sleep Mode testing, this display would qualify for the 0,7 W USB 3.x adder. Recalling that Formula, Formula.

Example 3: A computer monitor with one bridging and one network capability, USB 3.x and Wi-Fi, shall be tested with both capabilities connected and enabled during Sleep Mode testing. Assuming no additional capabilities are enabled during Sleep Mode testing, this display would qualify for the 0,7 W USB 3.x adder and the 2,0 W Wi-Fi adder. Recalling that Formula, Formula.

3.4.3.

For products that offer more than one Sleep Mode (e.g., ‘Sleep’ and ‘Deep Sleep’), measured Sleep Mode power (PSLEEP) in any Sleep Mode shall not exceed PSLEEP_MAX in the case of products without data or networking connection capabilities, or PSLEEP_AP, in the case of products tested with additional power-consuming capabilities, such as data bridge connections or networking connections. If the product has a variety of Sleep Modes that may be manually selected, or if the product can enter Sleep Mode via different methods (e.g., remote control or putting the host PC to sleep), the measured Sleep Mode power (PSLEEP) of the Sleep Mode with the highest PSLEEP, as measured per Section 6.5 of the Test Method, shall be the PSLEEP reported for qualification. If the product automatically transitions through its various Sleep Modes, the average PSLEEP of all Sleep Modes as measured in Section 6.5 of the Test Method shall be the PSLEEP reported for qualification

3.5.   Off Mode Requirements

Measured Off Mode power (POFF) shall be less than or equal to the Maximum Off Mode Power Requirement (POFF_MAX) specified in Table 5.

Table 5

Maximum Off Mode Power Requirement (POFF_MAX)

POFF_MAX

(watts)

0,5

3.6.   Maximum reported and maximum measured luminance shall be reported for all products; as shipped luminance shall be reported for all products except those with ABC enabled by default.

4.   Test requirements

4.1.   Test Methods

For products placed on the market of the European Union manufacturers are required to perform tests and self-certify those models that meet Energy Star guidelines. Test methods identified below shall be used to determine qualification for ENERGY STAR.

Product Type

Test Method

All Product Types and Screen Sizes

ENERGY STAR Test Method for Determining Displays Energy Use Version 6.0 – Rev. Jan-2013

4.2.   Number of Units Required for Testing

4.2.1.

One unit of a Representative Model, as defined in Section 1, shall be selected for testing.

4.2.2.

For qualification of a product family, the product configuration that represents the worst-case power consumption for each product category within the family shall be considered the Representative Model.

4.3.   International Market Qualification

Products shall be tested for qualification at the relevant input voltage/frequency combination for each market in which they will be sold and promoted as ENERGY STAR.

5.   User Interface

Manufacturers are encouraged to design products in accordance with the user interface standard, IEEE P1621: Standard for User Interface Elements in Power Control of Electronic Devices Employed in Office/Consumer Environments. For details, see http://eetd.LBL.gov/Controls. In the event that the manufacturer does not adopt IEEE P1621, the manufacturer shall provide EPA or the European Commission, as appropriate with its rationale for not doing so.

6.   Effective date

6.1.

The date that manufacturers may begin to qualify products as Energy Star under this Version 6.0, will be defined as the effective date of the Agreement. To qualify for ENERGY STAR, a product model shall meet the ENERGY STAR specification in effect on its date of manufacture. The date of manufacture is specific to each unit and is the date (e.g., month and year) on which a unit is considered to be completely assembled.

6.2.

Future Specification Revisions: EPA and the European Commission reserve the right to change this specification should technological and/or market changes affect its usefulness to consumers, industry, or the environment. In keeping with current policy, revisions to the specification are arrived at through stakeholder discussions. In the event of a specification revision, please note ENERGY STAR qualification is not automatically granted for the life of a model.

7.   Considerations for future revisions

7.1.   Displays Larger Than 61 in Diagonal Screen Size

It is understood that interactive displays greater than 60’ in diagonal screen size are currently available in the market and are namely used for commercial and educational purposes. There is interest in better understanding the power consumption associated with these products when tested according to the Displays Test Method and EPA and the European Commission will work with stakeholders prior to, and during, the next specification revision development process to access the information. EPA and the European Commission are in principle interested in exploring expanding the scope of products to those greater than 61’ in diagonal screen size in the next specification revision.

7.2.   Touch Screen Functionality

EPA and the European Commission are committed to continuing to develop performance levels for displays that account for new features and functionality, and anticipate that displays with touch screen functionality, which are included in the scope of this specification, will become more prevalent in the market, especially among signage displays. Going forward, EPA, DOE and the European Commission will explore with stakeholders whether touch screen functionality impacts On Mode power consumption to determine to what extent the next specification development process should address touch screen functionality.

II.   UNINTERRUPTIBLE POWER SUPPLIES SPECIFICATIONS

1.   Definitions

Unless otherwise specified, all terms used in this document are consistent with the definitions in the International Electrical Commission (IEC) standard IEC 62040-3 (1).

For the purpose of this specification the following definitions apply:

Uninterruptible Power Supply (UPS): Combination of convertors, switches, and energy storage devices (such as batteries) constituting a power system for maintaining continuity of load power in case of input power failure (2).

1.1.

Power conversion mechanism:

(a)

Static UPS: UPS where solid-state power electronic components provide the output voltage.

(b)

Rotary UPS: UPS where one or more electrical rotating machines provide the output voltage.

(1)

Rotary UPS (RUPS) without Diesel: A rotary UPS that does not contain an integral diesel engine to supply power to the load during an input power failure.

(2)

Diesel-coupled rotary UPS (DRUPS): A rotary UPS that contains an integral diesel engine that may be used to supply power to the load during an input power failure.

(c)

Power Output:

(1)

Alternating Current (Ac)-output UPS: UPS that supplies power with a continuous flow of electric charge that periodically reverses direction.

(2)

Direct Current (Dc)-output UPS/Rectifier: UPS that supplies power with a continuous flow of electric charge that is unidirectional. Includes both individual rectifier units for dc applications and entire dc-output UPS frames or systems, consisting of rectifier modules, controllers, and any other supporting components.

Note: Dc-output UPSs are also known as rectifiers. For the purposes of this document, the term ‘Dc-output UPS/Rectifier’ is used because a ‘rectifier’ may also refer to an Ac-output UPS subsystem.

1.2.

Modular UPS: A UPS comprised of two or more single UPS units, sharing one or more common frames and a common energy storage system, whose outputs, in Normal Mode of operation, are connected to a common output bus contained entirely within the frame(s). The total quantity of single UPS units in a modular UPS equals ‘n + r’ where n is the quantity of single UPS units required to support the load; r is the quantity of redundant UPS units. Modular UPSs may be used to provide redundancy, to scale capacity or both.

1.3.

Redundancy: Addition of UPS units in a parallel UPS to enhance the continuity of load power, and classified as follows.

(a)

N + 0: UPS that cannot tolerate any failures while maintaining Normal Mode operation. No redundancy.

(b)

N + 1: Parallel UPS that can tolerate the failure of one UPS unit or one group of UPS units while maintaining Normal Mode operation.

(c)

2N: Parallel UPS that can tolerate the failure of one half of its UPS units while maintaining Normal Mode operation.

1.4.

UPS Operational Modes:

(a)

Normal Mode: Stable mode of operation that the UPS attains under the following conditions:

(1)

Ac input supply is within required tolerances and supplies the UPS.

(2)

The energy storage system remains charged or is under recharge.

(3)

The load is within the specified rating of the UPS.

(4)

The Bypass is available and within specified tolerances (if applicable).

(b)

Stored Energy Mode: Stable mode of operation that the UPS attains under the following conditions:

(1)

Ac input power is disconnected or is out of required tolerance.

(2)

All power is derived from the energy storage system or, in the case of a DRUPS, from the integrated Diesel engine or a combination of both.

(3)

The load is within the specified rating of the UPS.

(c)

Bypass Mode: Mode of operation that the UPS attains when operating the load supplied via the Bypass only.

1.5.

UPS Input Dependency Characteristics:

(a)

Voltage and Frequency Dependent (VFD): Capable of protecting the load from power outage (3).

(b)

Voltage Independent (VI): Capable of protecting the load as required for VFD, above, and in addition from:

(1)

Under-voltage applied continuously to the input

(2)

Over-voltage applied continuously to the input (4)

(c)

Voltage and Frequency Independent (VFI): Independent of voltage and frequency variations and capable of protecting the load against adverse effects from such variations without depleting the stored energy source.

1.6.

Single-normal-mode UPS: A UPS that functions in Normal Mode within the parameters of only one set of input dependency characteristics. For example, a UPS that functions only as VFI.

1.7.

Multiple-normal-mode UPS: A UPS that functions in Normal Mode within the parameters of more than one set of input dependency characteristics. For example, a UPS that can function as either VFI or VFD.

1.8.

Bypass: Power path alternative to the ac converter.

(a)

Maintenance Bypass (path): Alternative power path provided to maintain continuity of load power during maintenance activities.

(b)

Automatic Bypass: Power path (primary or stand-by) alternative to the indirect ac converter.

(1)

Mechanical Bypass: control is via a switch with mechanically separable contacts

(2)

Static Bypass (electronic bypass): control is via an electronic power switch, for example transistors, thyristors, triacs or other semiconductor device or devices.

(3)

Hybrid Bypass: control is via switch with mechanically separable contacts in combination with at least one controlled electronic valve device.

1.9.

Reference Test Load: Load or condition in which the output of the UPS delivers the active power (W) for which the UPS is rated (5).

1.10.

Unit Under Test (UUT): The UPS undergoing the test, configured as though for shipment to the customer, and including any accessories (e.g., filters or transformers) necessary to meet the test setup as specified in Section 3 of the ENERGY STAR Test Method.

1.11.

Power Factor: Ratio of the absolute value of active power P to the apparent power S.

1.12.

Product Family: A group of product models that are (1) made by the same manufacturer, (2) subject to the same ENERGY STAR qualification criteria, and (3) of a common basic design. For UPSs, acceptable variations within a product family include:

(a)

Number of installed modules;

(b)

Redundancy;

(c)

Type and quantity of input and output filters;

(d)

Number of rectifier pulses (6); and

(e)

Energy storage system capacity.

1.13.

Abbreviations:

(a)   A: Ampere

(b)   ac: Alternating Current

(c)   dc: Direct Current

(d)   DRUPS: Diesel coupled rotary UPS

(e)   RUPS: Rotary UPS

(f)   THD: Total Harmonic Distortion

(g)   UPS: Uninterruptible Power Supply

(h)   UUT: Unit Under Test

(i)   V: Volt

(j)   VFD: Voltage and Frequency Dependent

(k)   VFI: Voltage and Frequency Independent

(l)   VI: Voltage Independent

(m)   W: Watt

(n)   Wh: Watt-hour

2.   Scope

2.1.   Products that meet the definition of an Uninterruptible Power Supply (UPS) as specified herein including Static and Rotary UPSs and Ac-output UPSs and Dc-output UPSs/Rectifiers are eligible for ENERGY STAR qualification, with the exception of products listed in Section 2.3.

2.2.   Products eligible for qualification under this specification include:

(a)

Consumer UPSs intended to protect desktop computers and related peripherals, and/or home entertainment devices such as TVs, set top boxes, DVRs, Blu-ray and DVD players;

(b)

Commercial UPSs intended to protect small business and branch office information and communication technology equipment such as servers, network switches and routers, and small storage arrays;

(c)

Data Center UPSs intended to protect large installations of information and communication technology equipment such as enterprise servers, networking equipment, and large storage arrays; and,

(d)

Telecommunications Dc-output UPSs/Rectifiers intended to protect telecommunication network systems located within a central office or at a remote wireless/cellular site.

2.3.   Excluded Products

2.3.1.

Products that are covered under other ENERGY STAR product specifications are not eligible for qualification under this specification. The list of specifications currently in effect can be found at www.eu-energystar.org.

2.3.2.

The following products are not eligible for qualification under this specification:

(a)

Products that are internal to a computer or another end-use load (e.g., battery supplemented internal power supplies or battery backup for modems, security systems, etc.);

(b)

Industrial UPSs specifically designed to protect critical control, manufacturing, or production processes or operations;

(c)

Utility UPSs designed for use as part of electrical transmission and distribution systems (e.g. electrical substation or neighbourhood-level UPSs);

(d)

Cable TV (CATV) UPSs designed to power the cable signal distribution system outside plant equipment and connected directly or indirectly to the cable itself. The ‘cable’ may be coaxial cable (metallic wire), fibre-optic, or wireless (e.g., ‘Wi-Fi’);

(e)

UPSs designed to comply with specific UL safety standards for safety-related applications, such as emergency lighting, operations or egress, or medical diagnostic equipment; and,

(f)

UPSs designed for mobile, shipboard, marine or airborne applications.

3.   Qualification Criteria

3.1.   Significant Digits and Rounding

3.1.1.

All calculations shall be carried out with directly measured (unrounded) values.

3.1.2.

Unless otherwise specified, compliance with specification limits shall be evaluated using directly measured or calculated values without any benefit from rounding.

3.1.3.

Directly measured or calculated values that are submitted for reporting on the ENERGY STAR website shall be rounded to the nearest significant digit as expressed in the corresponding specification limit.

3.2.   Energy Efficiency Requirements for Ac output UPSs

3.2.1.

Single-normal-mode UPSs: Average loading-adjusted efficiency (EffAVG), as calculated per Equation 1, shall be greater than or equal to the Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 2, for the specified rated output power and input dependency characteristic, except as specified below.

For products with rated output power greater than 10 000 W and communication and measurement capability, as specified in Section 3.6, average loading-adjusted efficiency (EffAVG), as calculated per Equation 1, shall be greater than or equal to the Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 3, for the specified input dependency characteristic.

Equation 1: Calculation of Average Efficiency for Ac-output UPSs

Formula

Where:

EffAVG is the average loading-adjusted efficiency,

tn % is the proportion of time spent at the particular n % of the Reference Test Load, as specified in the loading assumptions in Table 1, and

Eff|n % is the efficiency at the particular n % of the Reference Test Load, as measured according to the ENERGY STAR Test Method.

Table 1

Ac-output UPS Loading Assumptions for Calculating Average Efficiency

Rated Output Power, P,

in watts (W)

Input

Dependency

Characteristic

Proportion of Time Spent at Specified

Proportion of Reference Test Load,

tn %

25 %

50 %

75 %

100 %

P ≤ 1 500 W

VFD

0,2

0,2

0,3

0,3

VFD

0

0,3

0,4

0,3

1 500 W < P ≤ 10 000 W

VFD, VI, or VFI

0

0,3

0,4

0,3

P > 10 000 W

VFD, VI, or VFI

0,25

0,5

0,25

0


Table 2

Ac-output UPS Minimum Average Efficiency Requirement

Minimum Average Efficiency Requirement (EffAVG_MIN), where:

P is the Rated Output Power in watts (W), and

ln is the natural logarithm.

Rated Output Power

Input Dependency Characteristic

VFD

VI

VFI

P ≤ 1 500 W

0,967

0,0099 × ln(P) + 0,815

1 500 W < P ≤ 10 000 W

0,970

0,967

P > 10 000 W

0,970

0,950

0,0099 × ln(P) + 0,805


Table 3

Ac-output UPS Minimum Average Efficiency Requirement for Products with Metering and Communications Capability

Minimum Average Efficiency Requirement (EffAVG_MIN), where:

P is the Rated Output Power in watts (W), and

ln is the natural logarithm.

Rated Output Power

Input Dependency Characteristic

VFD

VI

VFI

P > 10 000 W

0,960

0,940

0,0099 × ln(P) + 0,795

3.2.2.

Multiple-normal-mode UPSs that Do Not Ship with the Highest Input Dependency Mode Enabled by Default: If the Multiple-normal-mode UPS does not ship with its highest input dependency mode enabled by default, its average loading-adjusted efficiency (EffAVG), as calculated per Equation 1, shall be greater than or equal to:

(a)

The Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 2, for the rated output power and lowest input dependency mode provided by the UPS, for models with output power less than or equal to 10 000 W or no communication and measurement capability as specified in Section 3.6; or

(b)

The Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 3, for the rated output power and lowest input dependency mode provided by the UPS, for models with output power greater than 10 000 W and communication and measurement capability as specified in Section 3.6.

3.2.3.

Multiple-normal-mode UPSs that Ship with the Highest Input Dependency Mode Enabled by Default: If the Multiple-normal-mode UPS does ship with its highest input dependency mode enabled by default, its average loading-adjusted efficiency (EffAVG), as calculated per Equation 2, shall be greater than or equal to:

(a)

The Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 2, for the rated output power and lowest input dependency mode provided by the UPS, for models with output power less than or equal to 10 000 W or no communication and measurement capability as specified in Section 3.6; or

(b)

The Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 3, for the rated output power and lowest input dependency mode provided by the UPS, for models with output power greater than 10 000 W and communication and measurement capability as specified in Section 3.6.

Equation 2: Calculation of Average Efficiency for Multiple-normal-mode Ac-output UPSs

Formula

Where:

EffAVG is the average loading adjusted efficiency,

Eff1 is the average loading adjusted efficiency in the lowest input dependency mode (i.e., VFI or VI), as calculated per Equation 1, and

Eff2 is the average loading adjusted efficiency in the highest input dependency mode (i.e., VFD), as calculated per Equation 1.

3.3.   Energy Efficiency Requirements for Dc-output UPSs/Rectifiers

Average loading-adjusted efficiency (EffAVG), as calculated per Equation 3, shall be greater than or equal to the Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 4. This requirement shall apply to complete systems and/or individual modules. Manufacturers can qualify either, subject to the following requirements:

(a)

Complete systems that are also modular shall be qualified as Modular UPS Product Families with a particular model of module installed,

(b)

Qualification of individual modules will have no bearing on the qualification of modular systems unless the entire systems are also qualified as specified above.

(c)

For products with rated output power greater than 10 000 W and communication and measurement capability, as specified in Section 3.6, average loading-adjusted efficiency (EffAVG), as calculated per Equation 3, shall be greater than or equal to the Minimum Average Efficiency Requirement (EffAVG_MIN), as determined per Table 5.

Equation 3: Calculation of Average Efficiency for All Dc-output UPSs

Formula

Table 4

Dc-output UPS/Rectifier Minimum Average Efficiency Requirement

Minimum Average Efficiency

Requirement (EffAVG_MIN)

0,955

Table 5

Dc-output UPS/Rectifier Minimum Average Efficiency Requirement for Products with Metering and Communications Capability

Rated Output

Power

Minimum Average Efficiency

Requirement (EffAVG_MIN)

P > 10 000 W

0,945

3.4.   Power Factor Requirements

The measured input power factor of all Ac-output UPSs at 100 percent of the Reference Test Load shall be greater than or equal to the Minimum Power Factor Requirement specified in Table 6 for all VFI and VI Normal Modes required for qualification.

Table 6

UPS Minimum Input Power Factor Requirement for Ac-output UPSs

Minimum Power Factor

Requirement

0,90

3.5.   Standard Information Reporting Requirements

3.5.1.

Data for a standardized Power and Performance Data Sheet (PPDS) shall be submitted to EPA and/or the European Commission for each model or Product Family.

3.5.2.

Further details on the PPDS can be found on the ENERGY STAR web page for UPS at www.energystar.gov/products.

The PPDS contains the following information:

(a)

General characteristics (manufacturer, model name and number);

(b)

Electrical characteristics (power conversion mechanism, topology, input and output voltage and frequency);

(c)

Average efficiency used for qualification;

(d)

Efficiency at each loading point and power factor test results, in each applicable Normal Mode, and for both the tested maximum and minimum configurations for Modular UPS Product Families;

(e)

Metering and communications ability (data displayed on the meter, data provided via the network, and available protocols);

(f)

Web link to an available public document containing model specific test procedure guidelines, if applicable;

(g)

Battery/stored energy device characteristics;

(h)

Physical dimensions

3.5.3.

EPA and the European Commission may periodically revise this PPDS, as necessary, and will notify Partners of the revision process.

3.6.   Communication and Measurement Requirements

3.6.1.

Ac-output UPSs and Dc-output UPSs/Rectifiers with rated output power greater than 10 000 W may qualify for a 1 percentage point efficiency incentive, as reflected in Tables 3 and 5, if sold with an energy meter possessing the following characteristics:

(a)

The meter is either shipped as an independent, external component bundled with the UPS at the point of sale or is integral to the UPS.

(b)

The meter measures UPS output energy in kWh in each Normal Mode.

(c)

The meter can communicate the measurement results over a network using one of the following protocols: Modbus RTU, Modbus TCP, or SNMP (v1, 2, or 3)

(d)

If the meter is external to the UPS, it meets the requirements in Section 3.6.2.

(e)

If the meter is integral to the UPS, it meets the requirements in Section 3.6.3.

3.6.2.

Requirements for External Meters: External meters bundled with the UPS shall meet one of the following requirements for the UPS to obtain the metering efficiency incentive:

(a)

Meet Accuracy Class 2 or better (i.e., Class 1, Class 0,5 S, or Class 0,2 S), as specified in IEC 62053-21 (7), IEC 62053-22 (8), or ANSI C12.2 (9);

(b)

Exhibit a relative error in energy measurement less than or equal to 2 percent compared to a standard under the conditions specified in Section 3.6.4, with the exception of current, which shall be tested at 25 percent and 100 percent of the meter’s maximum current; or

(c)

Exhibit a relative error in energy measurement less than or equal to 5 percent compared to a standard when part of a complete measurement system (including current transformers that could be integrated with the meter and UPS) under the conditions specified in Section 3.6.4.

3.6.3.

Requirements for Integral Meters: Integral meters shall meet the following requirements under the conditions specified in Section 3.6.4 for the UPS to obtain the metering efficiency incentive:

Exhibit a relative error in energy measurement less than or equal to 5 percent compared to a standard when part of a complete measurement system (including current transformers integrated with the meter and UPS).

3.6.4.

Environmental and Electrical Conditions for Meter Accuracy: The meter shall meet the requirements specified in Section 3.6.2 or 3.6.3 under the following conditions:

(a)

Environmental conditions: Consistent with the ENERGY STAR Test Method and the standards referenced therein; and

(b)

Electrical conditions: Consistent with each of the loading points in the ENERGY STAR Test Method and the standards referenced therein.

4.   Testing

4.1.   Test Methods

For products placed on the market of the European Union manufacturers are required to perform tests and self-certify those models that meet Energy Star guidelines. When testing UPSs, the test methods identified in Table 7 shall be used to determine ENERGY STAR qualification.

Table 7

Test Methods for ENERGY STAR Qualification

Product Type

Test Method

All UPSs

ENERGY STAR Test Method for Uninterruptible Power Supplies, Rev. May-2012

4.2.   Number of Units Required for Testing

4.2.1.

Representative Models shall be selected for testing per the following requirements:

(a)

For qualification of an individual product model, a product configuration equivalent to that which is intended to be marketed and labelled as ENERGY STAR is considered the Representative Model;

(b)

For qualification of a Modular UPS Product Family where models vary by number of installed modules, the manufacturer shall select the maximum and minimum configurations to serve as Representative Models—i.e., a modular system shall meet the eligibility criteria in both its maximum and minimum non-redundant configurations. If the maximum and minimum configuration Representative Models meet the ENERGY STAR qualification criteria at their respective output power levels, all intermediate configuration models within a Modular UPS Product Family may be qualified for ENERGY STAR.

(c)

For qualification of a UPS Product Family where the models are related by a characteristic other than the number of installed modules, the highest energy using configuration within the Product Family shall be considered the Representative Model with the exception of energy storage system variations—the manufacturer may select any energy storage system for the test, within the requirements of the ENERGY STAR Test Method. Other products within a Product Family do not have to be tested for qualification, but they are expected to meet relevant ENERGY STAR qualification criteria and may be subject to verification testing sometime after initial qualification.

4.2.2.

A single unit of each Representative Model shall be selected for testing.

4.2.3.

All tested units shall meet ENERGY STAR qualification criteria.

5.   Effective Date

5.1.

The date that manufacturers may begin to qualify products as Energy Star under this Version 1.0, will be defined as the effective date of the Agreement. To qualify for ENERGY STAR, a product model shall meet the ENERGY STAR specification in effect on its date of manufacture. The date of manufacture is specific to each unit and is the date on which a unit is considered to be completely assembled.

5.2.

Future Specification Revisions: EPA and the European Commission reserve the right to change this specification should technological and/or market changes affect its usefulness to consumers, industry, or the environment. In keeping with current policy, revisions to the specification are arrived at through stakeholder discussions. In the event of a specification revision, please note that the ENERGY STAR qualification is not automatically granted for the life of a product model.

III.   COMPUTER SERVERS SPECIFICATION (VERSION 2.0)

1.   Definitions

1.1.   Product Types

1.1.1.

Computer Server: A computer that provides services and manages networked resources for client devices (e.g., desktop computers, notebook computers, thin clients, wireless devices, PDAs, IP telephones, other computer servers, or other network devices). A computer server is sold through enterprise channels for use in data centres and office/corporate environments. A computer server is primarily accessed via network connections, versus directly-connected user input devices such as a keyboard or mouse. For purposes of this specification, a computer server must meet all of the following criteria:

(a)

is marketed and sold as a Computer Server;

(b)

is designed for and listed as supporting one or more computer server operating systems (OS) and/or hypervisors;

(c)

is targeted to run user-installed applications typically, but not exclusively, enterprise in nature;

(d)

provides support for error-correcting code (ECC) and/or buffered memory (including both buffered dual in-line memory modules (DIMMs) and buffered on board (BOB) configurations).

(e)

is packaged and sold with one or more ac-dc or dc-dc power supplies; and

(f)

is designed such that all processors have access to shared system memory and are visible to a single OS or hypervisor.

1.1.2.

Managed Server: A computer server that is designed for a high level of availability in a highly managed environment. For purposes of this specification, a managed server must meet all of the following criteria:

(a)

is designed to be configured with redundant power supplies; and

(b)

contains an installed dedicated management controller (e.g., service processor).

1.1.3.

Blade System: A system comprised of a blade chassis and one or more removable blade servers and/or other units (e.g., blade storage, blade network equipment). Blade systems provide a scalable means for combining multiple blade server or storage units in a single enclosure, and are designed to allow service technicians to easily add or replace (hot-swap) blades in the field.

(a)

Blade Server: A computer server that is designed for use in a blade chassis. A blade server is a high-density device that functions as an independent computer server and includes at least one processor and system memory, but is dependent upon shared blade chassis resources (e.g., power supplies, cooling) for operation. A processor or memory module that is intended to scale up a standalone server is not considered a Blade Server.

(1)

Multi-bay Blade Server: A blade server requiring more than one bay for installation in a blade chassis.

(2)

Single-wide Blade Server: A blade server requiring the width of a standard blade server bay.

(3)

Double-wide Blade Server: A blade server requiring twice the width of a standard blade server bay.

(4)

Half-height Blade Server: A blade server requiring one half the height of a standard blade server bay.

(5)

Quarter-height Blade Server: A blade server requiring one quarter the height of a standard server bay.

(6)

Multi-Node Blade Server: A blade server which has multiple nodes. The blade server itself is hot swappable, but the individual nodes are not.

(b)

Blade Chassis: An enclosure that contains shared resources for the operation of blade servers, blade storage, and other blade form-factor devices. Shared resources provided by a chassis may include power supplies, data storage, and hardware for dc power distribution, thermal management, system management, and network services.

(c)

Blade Storage: A storage device that is designed for use in a blade chassis. A blade storage device is dependent upon shared blade chassis resources (e.g., power supplies, cooling) for operation.

1.1.4.

Fully Fault Tolerant Server: A computer server that is designed with complete hardware redundancy, in which every computing component is replicated between two nodes running identical and concurrent workloads (i.e., if one node fails or needs repair, the second node can run the workload alone to avoid downtime). A fully fault tolerant server uses two systems to simultaneously and repetitively run a single workload for continuous availability in a mission critical application.

1.1.5.

Resilient Server: A computer server designed with extensive Reliability, Availability, Serviceability (RAS) and scalability features integrated in the micro architecture of the system, CPU and chipset. For purposes of ENERGY STAR qualification under this specification, a Resilient Server shall have the characteristics as described in Appendix B of this specification.

1.1.6.

Multi-node Server: A computer server that is designed with two or more independent server nodes that share a single enclosure and one or more power supplies. In a multi-node server, power is distributed to all nodes through shared power supplies. Server nodes in a multi-node server are not designed to be hot-swappable.

Dual-node Server: A common multi-node server configuration consisting of two server nodes.

1.1.7.

Server Appliance: A computer server that is bundled with a pre-installed OS and application software that is used to perform a dedicated function or set of tightly coupled functions. Server appliances deliver services through one or more networks (e.g., IP or SAN), and are typically managed through a web or command line interface. Server appliance hardware and software configurations are customized by the vendor to perform a specific task (e.g., name services, firewall services, authentication services, encryption services, and voice-over-IP (VoIP) services), and are not intended to execute user-supplied software.

1.1.8.

High Performance Computing (HPC) System: A computing system which is designed and optimized to execute highly parallel applications. HPC systems feature a large number of clustered homogeneous nodes often featuring high speed inter-processing interconnects as well as large memory capability and bandwidth. HPC systems may be purposely built, or assembled from more commonly available computer servers. HPC systems must meet ALL the following criteria:

(a)

Marketed and sold as a Computer Server optimized for higher performance computing applications;

(b)

Designed (or assembled) and optimized to execute highly parallel applications;

(c)

Consist of a number of typically homogeneous computing nodes, clustered primarily to increase computational capability;

(d)

Includes high speed inter-processing interconnections between nodes.

1.1.9.

Direct Current (dc) Server: A computer server that is designed solely to operate on a dc power source.

1.1.10.

Large Server: A resilient/scalable server which ships as a pre-integrated/pre-tested system housed in one or more full frames or racks and that includes a high connectivity I/O subsystem with a minimum of 32 dedicated I/O slots.

1.2.   Product Category

A second-order classification or sub-type within a product type that is based on product features and installed components. Product categories are used in this specification to determine qualification and test requirements.

1.3.   Computer Server Form Factors

1.3.1.

Rack-mounted Server: A computer server that is designed for deployment in a standard 19-inch data centre rack as defined by EIA-310, IEC 60297, or DIN 41494. For the purposes of this specification, a blade server is considered under a separate category and excluded from the rack-mounted category.

1.3.2.

Pedestal Server: A self-contained computer server that is designed with PSUs, cooling, I/O devices, and other resources necessary for stand-alone operation. The frame of a pedestal server is similar to that of a tower client computer.

1.4.   Computer Server Components

1.4.1.

Power Supply Unit (PSU): A device that converts ac or dc input power to one or more dc power outputs for the purpose of powering a computer server. A computer server PSU must be self-contained and physically separable from the motherboard and must connect to the system via a removable or hard-wired electrical connection.

(a)

Ac-Dc Power Supply: A PSU that converts line-voltage ac input power into one or more dc power outputs for the purpose of powering a computer server.

(b)

Dc-Dc Power Supply: A PSU that converts line-voltage dc input power to one or more dc outputs for the purpose of powering a computer server. For purposes of this specification, a dc-dc converter (also known as a voltage regulator) that is internal to a computer server and is used to convert a low voltage dc (e.g., 12 V dc) into other dc power outputs for use by computer server components is not considered a dc-dc power supply.

(c)

Single-output Power Supply: A PSU that is designed to deliver the majority of its rated output power to one primary dc output for the purpose of powering a computer server. Single-output PSUs may offer one or more standby outputs that remain active whenever connected to an input power source. For purposes of this specification, the total rated power output from any additional PSU outputs that are not primary and standby outputs shall be no greater than 20 watts. PSUs that offer multiple outputs at the same voltage as the primary output are considered single-output PSUs unless those outputs (1) are generated from separate converters or have separate output rectification stages, or (2) have independent current limits.

(d)

Multi-output Power Supply: A PSU that is designed to deliver the majority of its rated output power to more than one primary dc output for the purpose of powering a computer server. Multi-output PSUs may offer one or more standby outputs that remain active whenever connected to an input power source. For purposes of this specification, the total rated power output from any additional PSU outputs that are not primary and standby outputs is greater than or equal to 20 watts.

1.4.2.

I/O Device: A device which provides data input and output capability between a computer server and other devices. An I/O device may be integral to the computer server motherboard or may be connected to the motherboard via expansion slots (e.g., PCI, PCIe). Examples of I/O devices include discrete Ethernet devices, InfiniBand devices, RAID/SAS controllers, and Fibre Channel devices.

I/O Port: Physical circuitry within an I/O device where an independent I/O session can be established. A port is not the same as a connector receptacle; it is possible that a single connector receptacle can service multiple ports of the same interface.

1.4.3.

Motherboard: The main circuit board of the server. For purposes of this specification, the motherboard includes connectors for attaching additional boards and typically includes the following components: processor, memory, BIOS, and expansion slots.

1.4.4.

Processor: The logic circuitry that responds to and processes the basic instructions that drive a server. For purposes of this specification, the processor is the central processing unit (CPU) of the computer server. A typical CPU is a physical package to be installed on the server motherboard via a socket or direct solder attachment. The CPU package may include one or more processor cores.

1.4.5.

Memory: For purposes of this specification, memory is a part of a server external to the processor in which information is stored for immediate use by the processor.

1.4.6.

Hard Drive (HDD): The primary computer storage device which reads and writes to one or more rotating magnetic disk platters.

1.4.7.

Solid State Drive (SSD): A storage device that uses memory chips instead of rotating magnetic platters for data storage.

1.5.   Other Datacenter Equipment:

1.5.1.

Network Equipment: A device whose primary function is to pass data among various network interfaces, providing data connectivity among connected devices (e.g., routers and switches). Data connectivity is achieved via the routing of data packets encapsulated according to internet Protocol, Fibre Channel, InfiniBand or similar protocol.

1.5.2.

Storage Product: A fully-functional storage system that supplies data storage services to clients and devices attached directly or through a network. Components and subsystems that are an integral part of the storage product architecture (e.g., to provide internal communications between controllers and disks) are considered to be part of the storage product. In contrast, components that are normally associated with a storage environment at the data centre level (e.g., devices required for operation of an external SAN) are not considered to be part of the storage product. A storage product may be composed of integrated storage controllers, storage devices, embedded network elements, software, and other devices. While storage products may contain one or more embedded processors, these processors do not execute user-supplied software applications but may execute data-specific applications (e.g., data replication, backup utilities, data compression, install agents).

1.5.3.

Uninterruptible Power Supply (UPS): Combination of convertors, switches, and energy storage devices (such as batteries) constituting a power system for maintaining continuity of load power in case of input power failure.

1.6.   Operational Modes and Power States

1.6.1.

Idle State: The operational state in which the OS and other software have completed loading, the computer server is capable of completing workload transactions, but no active workload transactions are requested or pending by the system (i.e., the computer server is operational, but not performing any useful work). For systems where ACPI standards are applicable, Idle State correlates only to ACPI System Level S0.

1.6.2.

Active State: The operational state in which the computer server is carrying out work in response to prior or concurrent external requests (e.g., instruction over the network). Active state includes both (1) active processing and (2) data seeking/retrieval from memory, cache, or internal/external storage while awaiting further input over the network.

1.7.   Other Key Terms

1.7.1.

Controller System: A computer or computer server that manages a benchmark evaluation process. The controller system performs the following functions:

(a)

start and stop each segment (phase) of the performance benchmark;

(b)

control the workload demands of the performance benchmark;

(c)

start and stop data collection from the power analyser so that power and performance data from each phase can be correlated;

(d)

store log files containing benchmark power and performance information;

(e)

convert raw data into a suitable format for benchmark reporting, submission and validation; and

(f)

collect and store environmental data, if automated for the benchmark.

1.7.2.

Network Client (Testing): A computer or computer server that generates workload traffic for transmission to a unit under test (UUT) connected via a network switch.

1.7.3.

RAS Features: An acronym for reliability, availability, and serviceability features. RAS is sometimes expanded to RASM, which adds ‘Manageability’ criteria. The three primary components of RAS as related to a computer server are defined as follows:

(a)

Reliability Features: Features that support a server’s ability to perform its intended function without interruption due to component failures (e.g., component selection, temperature and/or voltage de-rating, error detection and correction).

(b)

Availability Features: Features that support a server’s ability to maximize operation at normal capacity for a given duration of downtime (e.g., redundancy [both at micro- and macro-level]).

(c)

Serviceability Features: Features that support a server’s ability to be serviced without interrupting operation of the server (e.g., hot plugging).

1.7.4.

Server Processor Utilization: The ratio of processor computing activity to full-load processor computing activity at a specified voltage and frequency, measured instantaneously or with a short term average of use over a set of active and/or idle cycles.

1.7.5.

Hypervisor: A type of hardware virtualization technique that enables multiple guest operating systems to run on a single host system at the same time.

1.7.6.

Auxiliary Processing Accelerators (APAs): Computing expansion add-in cards installed in general-purpose add-in expansion slots (e.g., GPGPUs installed in a PCI slot).

1.7.7.

Buffered DDR Channel: Channel or Memory Port connecting a Memory Controller to a defined number of memory devices (e.g. DIMMs) in a computer server. A typical computer server may contain multiple Memory Controllers, which may in turn support one or more Buffered DDR Channels. As such, each Buffered DDR Channel serves only a fraction of the total addressable memory space in a computer server.

1.8.   Product Family

A high-level description referring to a group of computers sharing one chassis/motherboard combination that often contains hundreds of possible hardware and software configurations.

1.8.1.   Common Product Family Attributes: A set of features common to all models/configurations within a product family that constitute a common basic design. All models/configurations within a product family must share the following:

(a)

Be from the same model line or machine type;

(b)

Either share the same form factor (i.e., rack-mounted, blade, pedestal) or share the same mechanical and electrical designs with only superficial mechanical differences to enable a design to support multiple form factors;

(c)

Either share processors from a single defined processor series or share processors that plug into a common socket type.

(d)

Share PSUs that perform with efficiencies greater than or equal to the efficiencies at all required load points specified in Section 3.2 (i.e., 10 %, 20 %, 50 %, and 100 % of maximum rated load for single-output; 20 %, 50 %, and 100 % of maximum rated load for multi-output).

1.8.2.   Product Family Tested Product Configurations

(a)

Purchase Consideration Variations:

(1)

Low-end Performance Configuration: The combination of Processor Socket Power, PSUs, Memory, Storage (HDD/SDD), and I/O devices that represents the lower-price or lower-performance computing platform within the Product Family.

(2)

High-end Performance Configuration: The combination of Processor Socket Power, PSUs, Memory, Storage (HDD/SDD), and I/O devices that represents either the higher-price or higher-performance computing platform within the Product Family.

(b)

Typical Configuration:

Typical Configuration: A product configuration that lies between the Minimum and Maximum Power configurations and is representative of a deployed product with high volume sales.

(c)

Power Utilization Variations:

(1)

Minimum Power Configuration: The minimum configuration that is able to boot and execute supported OSs. The Minimum Configuration contains the lowest Processor Socket Power, least number of installed PSUs, Memory, Storage (HDD/SDD), and I/O devices, that is both offered for sale and capable of meeting ENERGY STAR requirements.

(2)

Maximum Power Configuration: The vendor-selected combination of components that maximize power usage within the Product Family once assembled and operated. The Maximum Configuration contains the highest Processor Socket Power, greatest number of installed PSUs, Memory, Storage (HDD/SDD), and I/O devices that is both offered for sale and capable of meeting ENERGY STAR requirements.

2.   Scope

2.1.   Included Products

A product must meet the definition of a Computer Server provided in Section 1 of this document to be eligible for ENERGY STAR qualification under this specification. Eligibility under Version 2.0 is limited to Blade-, Multi-node, Rack-mounted, or Pedestal form factor computer servers with no more than four processor sockets in the computer server (or per blade or node in the case of blade or multi-node servers) Products explicitly excluded from Version 2.0 are identified in Section 2.2.

2.2.   Excluded Products

2.2.1.

Products that are covered under other ENERGY STAR product specifications are not eligible for qualification under this specification. The list of specifications currently in effect can be found at www.eu-energystar.org/.

2.2.2.

The following products are not eligible for qualification under this specification:

(a)

Fully Fault Tolerant Servers;

(b)

Server Appliances;

(c)

High Performance Computing Systems;

(d)

Large Servers;

(e)

Storage Products including Blade Storage; and

(f)

Network Equipment.

3.   Qualification Criteria

3.1.   Significant Digits and Rounding

3.1.1.

All calculations shall be carried out with directly measured (unrounded) values.

3.1.2.

Unless otherwise specified, compliance with specification limits shall be evaluated using directly measured or calculated values without any benefit from rounding.

3.1.3.

Directly measured or calculated values that are submitted for reporting on the ENERGY STAR website shall be rounded to the nearest significant digit as expressed in the corresponding specification limit.

3.2.   Power Supply Requirements

3.2.1.

Power supply test data and test reports from testing entities recognized by EPA to perform power supply testing shall be accepted for the purpose of qualifying the ENERGY STAR product.

3.2.2.

Power Supply Efficiency Criteria: Power Supplies used in products eligible under this specification must meet the following requirements when tested using the Generalized Internal Power Supply Efficiency Test Protocol, Rev. 6.6 (available at www.efficientpowersupplies.org). Power Supply data generated using Rev. 6.4.2 (as required in Version 1.1), 6.4.3, or 6.5 are acceptable provided the test was conducted prior to the effective date of Version 2.0 of this specification.

(a)

Pedestal and Rack-mounted Servers: To qualify for ENERGY STAR, a pedestal or rack-mounted computer server must be configured with only PSUs that meet or exceed the applicable efficiency requirements specified in Table 1 prior to shipment.

(b)

Blade and Multi-node Servers: To qualify for ENERGY STAR, a Blade or Multi-node computer server shipped with a chassis must be configured such that all PSUs supplying power to the chassis meet or exceed the applicable efficiency requirements specified in Table 1 prior to shipment.

Table 1

Efficiency Requirements for PSUs

Power Supply Type

Rated Output Power

10 % Load

20 % Load

50 % Load

100 % Load

Multi-output

(Ac-Dc)

All Output Levels

N/A

85 %

88 %

85 %

Single-output

(Ac-Dc)

All Output Levels

80 %

88 %

92 %

88 %

3.2.3.

Power Supply Power Factor Criteria: Power Supplies used in Computers eligible under this specification must meet the following requirements when tested using the Generalized Internal Power Supply Efficiency Test Protocol, Rev. 6.6 (available at www.efficientpowersupplies.org). Power Supply data generated using Rev. 6.4.2 (as required in Version 1.1), 6.4.3, or 6.5 are acceptable provided the test was conducted prior to the effective date of Version 2.0.

(a)

Pedestal and Rack-mounted Servers: To qualify for ENERGY STAR, a pedestal or rack-mounted computer server must be configured with only PSUs that meet or exceed the applicable power factor requirements specified in Table 2 prior to shipment, under all loading conditions for which output power is greater than or equal to 75 watts. Partners are required to measure and report PSU power factor under loading conditions of less than 75 watts, though no minimum power factor requirements apply.

(b)

Blade or Multi-node Servers: To qualify for ENERGY STAR, a Blade or Multi-node computer server shipped with a chassis must be configured such that all PSUs supplying power to the chassis meet or exceed the applicable power factor requirements specified in Table 2 prior to shipment, under all loading conditions for which output power is greater than or equal to 75 watts. Partners are required to measure and report PSU power factor under loading conditions of less than 75 watts, though no minimum power factor requirements apply.

Table 2

Power Factor Requirements for PSUs

Power Supply Type

Rated Output Power

10 % Load

20 % Load

50 % Load

100 % Load

Ac-Dc Multi-output

All Output Ratings

N/A

0,80

0,90

0,95

Ac-Dc Single-output

Output Rating ≤ 500 W

N/A

0,80

0,90

0,95

Output Rating > 500 W

and

Output Rating ≤ 1 000 W

0,65

0,80

0,90

0,95

Output Rating > 1 000 watts

0,80

0,90

0,90

0,95

3.3.   Power Management Requirements

3.3.1.

Server Processor Power Management: To qualify for ENERGY STAR, a Computer Server must offer processor power management that is enabled by default in the BIOS and/or through a management controller, service processor, and/or the operating system shipped with the computer server. All processors must be able to reduce power consumption in times of low utilization by:

(a)

reducing voltage and/or frequency through Dynamic Voltage and Frequency Scaling (DVFS), or

(b)

enabling processor or core reduced power states when a core or socket is not in use.

3.3.2.

Supervisor Power Management: To qualify for ENERGY STAR, a product which offers a pre-installed supervisor system (e.g., operating system, hypervisor) must offer supervisor system power management that is enabled by default.

3.3.3.

Power Management Reporting: To qualify for ENERGY STAR, all power management techniques that are enabled by default must be itemized on the Power and Performance Data Sheet. This requirement applies to power management features in the BIOS, operating system, or any other origin that can be configured by the end-user.

3.4.   Blade and Multi-Node System Criteria

3.4.1.

Blade and Multi-Node Thermal Management and Monitoring: To qualify for ENERGY STAR, a blade or multi-node server must provide real-time chassis or blade/node inlet temperature monitoring and fan speed management capability that is enabled by default.

3.4.2.

Blade and Multi-Node Server Shipping Documentation: To qualify for ENERGY STAR, a blade or multi-node server that is shipped to a customer independent of the chassis must be accompanied with documentation to inform the customer that the blade or multi-node server is ENERGY STAR qualified only if it is installed in a chassis meeting requirements in Section 3.4.1 of this document. A list of qualifying chassis and ordering information must also be provided as part of product collateral provided with the blade or multi-node server. These requirements may be met via either printed materials, electronic documentation provided with the blade or multi-node server, or information publically available on the Partner’s website where information about the blade or multi-node server is found.

3.5.   Active State Efficiency Criteria

3.5.1.

Active State Efficiency Reporting: To qualify for ENERGY STAR, a Computer Server or Computer Server Product Family must be submitted for qualification with the following information disclosed in full and in the context of the complete Active State efficiency rating test report:

(a)

Final SERT rating tool results, which include the results files (both html and text format) and all results-chart png files; and

(b)

Intermediate SERT rating tool results over the entire test run, which include the results-details files (both html and text format) and all results-details-chart png files.

Data reporting and formatting requirements are discussed in Section 4.1 of this specification.

3.5.2.

Incomplete Reporting: Partners shall not selectively report individual workload module results, or otherwise present efficiency rating tool results in any form other than a complete test report, in customer documentation or marketing materials.

3.6.   Idle State Efficiency Criteria – One-Socket (1S) and Two-Socket (2S) Servers (neither Blade nor Multi-Node)

3.6.1.

Idle State Data Reporting: Maximum Idle State power (PIDLE_MAX) shall be measured and reported, both in qualification materials and as required in Section 4.

3.6.2.

Idle State Efficiency: Measured Idle State power (PIDLE) shall be less than or equal to the Maximum Idle State Power Requirement (PIDLE_MAX), as calculated per Equation 1.

Equation 1: Calculation of Maximum Idle State Power

Formula

Where:

PIDLE_MAX is the Maximum Idle State Power Requirement,

PBASE is the base idle power allowance, as determined per Table 3,

PADDL_i is the Idle State power allowance for additional components, as determined per Table 4.

(a)

These Idle power limits are applicable to one and two socket systems only.

(b)

Use Section 6.1 of the ENERGY STAR Computer Servers Test Method to determine the Idle State power for qualification.

(c)

The Resilient category in Table 3 applies only to two socket systems that meet the definition of Resilient Server as set forth in Appendix B.

(d)

All quantities (with the exception of installed processors) in Tables 3 and 4 refer to the number of components installed in the system, not the maximum number of components the system can support (e.g., installed memory, not supported memory; etc.)

(e)

The Additional Power Supply allowance may be applied for each redundant power supply used in the configuration.

(f)

For the purposes of determining Idle power allowances, all memory capacities shall be rounded to the nearest GB (10).

(g)

The Additional I/O Device allowance may be applied for all I/O Devices over the Base Configuration (i.e., Ethernet devices additional to two ports greater than or equal to 1 Gigabit per second (Gbit/s), onboard Ethernet, plus any non-Ethernet I/O devices), including on-board I/O devices and add-in I/O devices installed through expansion slots. This allowance may be applied for each of the following types of I/O functionality: Ethernet, SAS, SATA, Fibre Channel and Infiniband.

(h)

The Additional I/O Device allowance shall be calculated based upon the rated link speed of a single connection, rounded to the nearest Gbit. I/O devices with less than 1 Gbit speed do not qualify for the Additional I/O Device allowance.

(i)

The Additional I/O Device allowance shall only be applied for I/O devices that are active/enabled upon shipment, and are capable of functioning when connected to an active switch.

Table 3

Base Idle State Power Allowances for 1S and 2S Servers

Category

Maximum Possible Number of Installed Processors

(# P)

Managed Server

Base Idle State Power Allowance, PBASE

(watts)

A

1

No

47,0

B

1

Yes

57,0

C

2

No

92,0

D

2

Yes

142,0

Resilient

2

Yes

205,0


Table 4

Additional Idle Power Allowances for Extra Components

System Characteristic

Applies To:

Additional Idle Power Allowance

Additional

Power Supplies

Power supplies installed explicitly for power redundancy

20 watts per Power Supply

Hard Drives (including solid state drives)

Per installed hard drive

8,0 watts per Hard Drive

Additional Memory

Installed memory greater than 4 GB

0,75 watts per GB

Additional Buffered DDR Channel

Installed buffered DDR Channels greater than 8 channels

(Resilient Servers only)

4,0 watts per Buffered DDR Channel

Additional

I/O Devices

Installed Devices greater than two ports of ≥ 1 Gbit, onboard Ethernet

< 1 Gbit: No Allowance

= 1 Gbit: 2,0 watts/Active Port

> 1 Gbit and < 10 Gbit: 4,0 watts/Active Port

≥ 10 Gbit: 8,0 watts/Active Port

3.7.   Idle State Efficiency Criteria – Three-Socket (3S) and Four-Socket (4S) Servers (neither Blade nor Multi-Node)

Idle State Data Reporting: Idle State power (PIDLE) shall be measured and reported, both in qualification materials and as required in Section 4.

3.8.   Idle State Efficiency Criteria – Blade Servers

3.8.1.

Idle State Data Reporting: Idle State power (PTOT_BLADE_SYS) and (PBLADE) shall be measured and reported, both in qualification materials and as required in Section 4.

3.8.2.

The testing of Blade Servers for compliance with Section 3.8.1 shall be carried out under all of the following conditions:

(a)

Power values shall be measured and reported using a half-populated Blade Chassis. Blade Servers with multiple power domains, choose the number of power domains that is closest to filling half of the Blade Chassis. In a case where there are two choices that are equally close to half, test with the domain or combination of domains which utilize a higher number of Blade Servers. The number of blades tested during the half-populated Blade Chassis test shall be reported.

(b)

Power for a fully-populated blade chassis may be optionally measured and reported, provided that half-populated chassis data is also provided.

(c)

All Blade Servers installed in the Blade Chassis shall share the same configuration (homogeneous).

(d)

Per-blade power values shall be calculated using Equation 2.

Equation 2: Calculation of Single Blade Power

Formula

Where:

PBLADE is the per-Blade Server Power

PTOT_BLADE_SYS is total measured power of the Blade System,

NINST_BLADE_SRV is the number of installed Blade Servers in the tested Blade Chassis.

3.9.   Idle State Efficiency Criteria – Multi-Node Servers

3.9.1.

Idle State Data Reporting: Idle State power (PTOT_NODE_SYS) and (PNODE) shall be measured and reported, both in qualification materials and as required in Section 4, below.

3.9.2.

The testing of Multi-Node Servers for compliance with Section 3.9.1 shall be carried out under all of the following conditions:

(a)

Power values shall be measured and reported using a fully-populated Multi-Node Chassis.

(b)

All Multi-Node Servers in the Multi-Node Chassis shall share the same configuration (homogeneous).

(c)

Per-node power values shall be calculated using Equation 3.

Equation 3: Calculation of Single Node Power

Formula

Where:

PNODE is the per Node Server Power

PTOT_NODE_SYS is total measured power of the Multi Node Server,

NINST_NODE_SRV is the number of installed Multi Node Servers in the tested Multi-Node Chassis.

3.10.   Other Testing Criteria

APA Requirements: For all computer servers sold with APAs, the following criteria and provisions apply:

(a)

For single configurations: All Idle State testing shall be conducted both with and without the APAs installed. Idle Power measurements taken both with the APAs installed and removed shall be submitted to EPA or the European Commission, as appropriate as part of ENERGY STAR qualification materials.

(b)

For Product Families: Idle State testing shall be conducted both with and without the APAs installed in the Maximum Power/High-end Performance Configuration found in 1.8.2. Testing with and without the APAs installed may optionally be conducted and disclosed at the other test points.

(c)

Idle State power measurements taken both with the APAs installed and removed shall be submitted to EPA or the European Commission, as appropriate as part of ENERGY STAR qualification materials. These measurements shall be submitted for each individual APA product that is intended for sale with the qualified configuration.

(d)

Measurements of PIDLE in Sections 3.6 and 3.7, PBLADE in Section 3.8 and PNODE in Section 3.9 shall be performed with APAs removed, even if they are installed as-shipped. These measurements shall then be repeated with each APA installed, one at a time, to evaluate Idle State power consumption of each installed APA.

(e)

The Idle State power consumption of each installed APA in qualified configurations shall not exceed 46 watts.

(f)

The Idle State power consumption of each individual APA product sold with a qualified configuration shall be reported.

4.   Standard Information Reporting Requirements

Data Reporting Requirements

4.1.

All required data fields in the ENERGY STAR Version 2.0 Computer Servers Qualified Product Exchange form shall be submitted to the European Commission for each ENERGY STAR qualified Computer Server or Computer Server Product Family.

(a)

Partners are encouraged to provide one set of data for each ENERGY STAR qualified product configuration, though the European Commission will also accept a data set for each qualified product family.

(b)

A product family qualification must include data for all defined test points in 1.8.2, as applicable.

(c)

Whenever possible, Partners must also provide a hyperlink to a detailed power calculator on their Web site that purchasers can use to understand power and performance data for specific configurations within the product family.

4.2.

The following data will be displayed on the EU ENERGY STAR Web site through the product finder tool:

(a)

model name and number, identifying SKU and/or configuration ID;

(b)

system characteristics (form factor, available sockets/slots, power specifications, etc.);

(c)

system type (unmanaged, managed, scalable, etc.);

(d)

system configuration(s) (including Low-end Performance Configuration, High-end Performance Configuration, Minimum Power Configuration, Maximum Power Configuration, and Typical Configuration for Product Family qualification);

(e)

power consumption and performance data from required Active and Idle State Efficiency Criteria testing including results.xml, results.html, results.txt, all results-chart png files, results-details.html, results-details.txt, all results-details-chart png files;

(f)

available and enabled power saving features (e.g., power management);

(g)

a list of selected data from the ASHRAE Thermal Report;

(h)

inlet air temperature measurements made prior to the start of testing, at the conclusion of Idle State testing, and at the conclusion of Active State testing;

(i)

for product family qualifications, a list of qualified configurations with qualified SKUs or configuration IDs; and

(j)

for a blade server, a list of compatible blade chassis that meet ENERGY STAR qualification criteria.

4.3.

EPA and the European Commission may periodically revise this list, as necessary, and will notify and invite stakeholder engagement in such a revision process.

5.   Standard Performance Data Measurement and Output Requirements

5.1.   Measurement and Output

5.1.1.

A computer server must provide data on input power consumption (W), inlet air temperature (°C), and average utilization of all logical CPUs. Data must be made available in a published or user-accessible format that is readable by third-party, non-proprietary management software over a standard network. For blade and multi-node servers and systems, data may be aggregated at the chassis level.

5.1.2.

Computer servers classified as Class B equipment as set out in EN 55022:2006 are exempt from the requirements to provide data on input power consumption and inlet air temperature in 5.1.1. Class B refers to household and home office equipment (intended for use in the domestic environment). All computer servers in the program must meet the requirement and conditions to report utilization of all logical CPUs.

5.2.   Reporting Implementation

5.2.1.

Products may use either embedded components or add-in devices that are packaged with the computer server to make data available to end users (e.g., a service processor, embedded power or thermal meter (or other out-of-band technology), or pre-installed OS);

5.2.2.

Products that include a pre-installed OS must include all necessary drivers and software for end users to access standardized data as specified in this document. Products that do not include a pre-installed OS must be packaged with printed documentation of how to access registers that contain relevant sensor information. This requirement may be met via either printed materials, electronic documentation provided with the computer server, or information publically available on the Partner’s website where information about the computer server is found.

5.2.3.

When an open and universally available data collection and reporting standard becomes available, manufacturers should incorporate the universal standard into their systems;

5.2.4.

Evaluation of the accuracy (5.3) and sampling (5.4) requirements shall be completed through review of data from component product datasheets. If this data is absent, Partner declaration shall be used to evaluate accuracy and sampling.

5.3.   Measurement Accuracy

5.3.1.

Input power: Measurements must be reported with accuracy of at least ± 5 % of the actual value, with a maximum level of accuracy of ± 10 W for each installed PSU (i.e., power reporting accuracy for each power supply is never required to be better than ± 10 watts) through the operating range from Idle to full power;

5.3.2.

Processor utilization: Average utilization must be estimated for each logical CPU that is visible to the OS and must be reported to the operator or user of the computer server through the operating environment (OS or hypervisor);

5.3.3.

Inlet air temperature: Measurements must be reported with an accuracy of at least ± 2 °C.

5.4.   Sampling Requirements

5.4.1.

Input power and processor utilization: Input power and processor utilization measurements must be sampled internally to the computer server at a rate of greater than or equal to measurement per contiguous 10 second period. A rolling average, encompassing a period of no more than 30 seconds, must be sampled internally to the computer server at a frequency of greater than or equal to once per ten seconds.

5.4.2.

Inlet air temperature: Inlet air temperature measurements must be sampled internally to the computer server at a rate of greater than or equal to 1 measurement every 10 seconds.

5.4.3.

Time stamping: Systems that implement time stamping of environmental data shall sample internally to the computer server data at a rate of greater than or equal to 1 measurement every 30 seconds.

5.4.4.

Management Software: All sampled measurements shall be made available to external management software either via an on-demand pull method, or via a coordinated push method. In either case the system’s management software is responsible for establishing the data delivery time scale while the computer server is responsible to assuring data delivered meets the above sampling and accuracy requirements.

6.   Testing

6.1.   Test Methods

6.1.1.

When testing Computer Server products, the test methods identified in Table 5 shall be used to determine ENERGY STAR qualification.

Table 5

Test Methods for ENERGY STAR Qualification

Product Type or Component

Test Method

All

ENERGY STAR Test Method for Computer Servers (Rev. March-2013)

All

Standard Performance Evaluation Corporation (SPEC) Server Efficiency Rating Tool (SERT), Version 1.0.0, Rev. Feb 26, 2013

6.1.2.

When testing Computer Server products, UUTs must have all Processor Sockets populated during testing.

If a Computer Server cannot support populating all Processor Sockets during testing, then the system must be populated to its maximum functionality. These systems will be subject to the base idle state power allowance based on the number of sockets in the system.

6.2.   Number of Units Required for Testing

Representative Models shall be selected for testing per the following requirements:

(a)

For qualification of an individual product configuration, the unique configuration that is intended to be marketed and labelled as ENERGY STAR is considered the Representative Model.

(b)

For qualification of a product family of all product types, one product configuration for each of the five points identified in definitions 1.8.2 within the family are considered Representative Models. All such representative models shall have the same Common Product Family Attributes as defined in 1.8.1.

6.3.   Qualifying Families of Products

6.3.1.

Partners are encouraged to test and submit data on individual product configurations for qualification to ENERGY STAR. However, a Partner may qualify multiple product configurations under one Product Family designation if each configuration within the family meets one of the following requirements:

(a)

Individual products are built on the same platform, are eligible under and meet the same specific requirements in this specification, and are identical in every respect to the tested, representative product configuration except for housing and colour; or

(b)

Individual products meet the requirements of a product family, as defined in Section 1.8, above. In this case, partners must test and submit data as required in Section (b).

6.3.2.

Partners are required to submit a Power and Performance Data Sheet for each product family that is submitted for qualification.

6.3.3.

All product configurations within a product family that is submitted for qualification must meet ENERGY STAR requirements, including products for which data was not reported.

7.   Effective Date

7.1.

The effective date of this Version 2.0 ENERGY STAR Computer Servers specification will be defined as the effective date of the Agreement. To qualify for ENERGY STAR, a product model shall meet the ENERGY STAR specification in effect on its date of manufacture. The date of manufacture is specific to each unit and is the date on which a unit is considered to be completely assembled.

7.2.

Future Specification Revisions: EPA and the European Commission reserve the right to change this specification should technological and/or market changes affect its usefulness to consumers, industry, or the environment. In keeping with current policy, revisions to the specification are arrived at through stakeholder discussions. In the event of a specification revision, please note that the ENERGY STAR qualification is not automatically granted for the life of a product model.

8.   Considerations for Future Revisions

8.1.

Active State Efficiency Criteria: EPA and the European Commission intend to set active state efficiency criteria in Version 3.0 for all computer server categories in which it has enough SERT data to adequately differentiate products.

8.2.

Right Sizing of Power Supplies: EPA and the European Commission will investigate opportunities for encouraging right-sizing of power supplies in Version 3.0.

8.3.

Inclusion of Dc-Dc Computer Servers: EPA and the European Commission encourage manufacturers to work with SPEC to develop support for dc servers in the SERT, so that dc computer servers may be considered for qualification in Version 3.0.

8.4.

Inclusion of Additional System Architectures: EPA and the European Commission encourage manufacturers to work with SPEC to develop support for architectures that are not currently supported by the SERT, but which represent a sizeable portion of the Computer Servers market. EPA and the European Commission will consider any architecture that is supported by the SERT prior to the development of Version 3.0.

8.5.

Removal of Adder for Additional Redundant Power Supplies: EPA and the European Commission are aware of technology that allows redundant power supplies to be kept in standby mode and only activated when needed. EPA and the European Commission encourage the adoption of this technology in computer servers, and will investigate whether the current adder for additional redundant power supplies is still necessary in Version 3.0.

8.6.

Auxiliary Processing Accelerator (APA) Requirements: EPA and the European Commission intends to revisit and potentially expand APA requirements in Version 3.0, based on APA data collected from Version 2.0 as well as the potential incorporation of APA evaluation in the SERT.

8.7.

Thermal Reporting and Testing Requirements: EPA and the European Commission plans to re-evaluate current temperature reporting and testing requirements to maximize the value of the data collected for manufacturers as well as data centre operators.

Appendix A

Sample Calculations

1.   Idle State Power Requirements

To determine the Maximum Idle State Power Requirement for ENERGY STAR qualification, determine the base idle state level from Table 3, and then add power allowances from Table 4 (provided in Section 3.6 of this Eligibility Criteria). An example is provided below:

Example: A standard single processor Computer Server with 8 GB of memory, two hard drives, and two I/O devices (the first with two 1 Gbit ports and the second with six 1 Gbit ports).

1.1.

Base allowance:

(a)

Determine base idle allowance from Table 3, provided for reference below.

(b)

The example server is evaluated under Category A and could consume no more than 47,0 watts in Idle to qualify for ENERGY STAR.

Category

Number of Installed Processors

(# P)

Managed Server

Base Idle Power Allowance

(W)

A

1

No

47,0

B

1

Yes

57,0

C

2

No

92,0

D

2

Yes

142,0

Resilient

2

Yes

205,0

1.2.

Additional Idle Power Allowances: Calculate additional idle allowances for extra components from Table 4, provided for reference below.

System Characteristic

Applies To

Additional Idle Power Allowance

Additional Power Supplies

Power supplies installed explicitly for power redundancy

20,0 watts per Power Supply

Hard Drives (including solid state drives)

All installed hard drives

8,0 watts per Hard Drive

Additional Memory

Installed memory greater than 4 GB

0,75 watts per GB

Additional Buffered DDR Channel

Installed buffered DDR Channels greater than 8 channels

(Resilient Servers only)

4,0 watts per Buffered DDR Channel

Additional I/O Devices (single connection speed rounded to nearest Gbit)

Installed Devices greater than two ports of 1 Gbit, onboard Ethernet

< 1 Gbit: No Allowance

= 1 Gbit: 2,0 watts/Active Port

> 1 Gbit and < 10 Gbit: 4,0 watts/Active Port

≥ 10 Gbit: 8,0 watts/Active Port

(a)

The example server has two hard drives. It therefore is provided with an additional 16,0 watt allowance for each hard drive (2 HDD × 8,0 watts).

(b)

The example server has 4 GB in excess of the base configuration. It therefore is provided with an additional 3,0 watt allowance for memory (4 extra GB × 0,75 watts/GB).

(c)

The example server has one I/O card that does not qualify for an adder: the first device has only two Ethernet ports and does not exceed the two-port threshold. Its second device does qualify for an adder: the server is provided with an additional 12,0 watt allowance for the device (six 1Gbit ports × 2,0 watts/active port).

1.3.

Calculate the final idle allowance by adding the base allowance with the additional power allowances. The example system would be expected to consume no more than 78,0 watts at Idle to qualify (47,0 W + 16,0 W + 3,0 W + 12,0 W).

2.   Additional Idle Allowance — Power Supplies

The following examples illustrate the idle power allowances for additional power supplies:

2.1.

If a Computer Server requires two power supplies to operate, and the configuration includes three installed power supplies, the server would receive an additional 20,0 watt idle power allowance.

2.2.

If the same server were instead shipped with four installed power supplies, it would receive an additional idle power allowance of 40,0 watts.

3.   Additional Idle Allowance — Additional Buffered DDR Channel

The following examples illustrate the idle power allowances for additional buffered DDR channels:

3.1.

If a resilient Computer Server is shipped with six installed buffered DDR channels, the server would not receive an additional idle power allowance.

3.2.

If the same resilient server were instead shipped with 16 installed buffered DDR channels, it would receive an additional idle power allowance of 32,0 watts (first 8 channels = no additional allowance, second 8 channels = 4,0 watts × 8 buffered DDR channels).

Appendix B

Identifying resilient server class

1.

Processor RAS and Scalability — All of the following shall be supported:

1.1.

Processor RAS: The processor must have capabilities to detect, correct, and contain data errors, as described by all of the following:

(a)

Error detection on L1 caches, directories and address translation buffers using parity protection;

(b)

Single bit error correction (or better) using ECC on caches that can contain modified data. Corrected data is delivered to the recipient (i.e., error correction is not used just for background scrubbing);

(c)

Error recovery and containment by means of (1) processor checkpoint retry and recovery, (2) data poison indication (tagging) and propagation, or (3) both. The mechanisms notify the OS or hypervisor to contain the error within a process or partition, thereby reducing the need for system reboots; and

(d)

(1) Capable of autonomous error mitigation actions within processor hardware, such as disabling of the failing portions of a cache, (2) support for predictive failure analysis by notifying the OS, hypervisor, or service processor of the location and/or root cause of errors, or (3) both.

1.2.

The processor technology used in resilient and scalable servers is designed to provide additional capability and functionality without additional chipsets, enabling them to be designed into systems with 4 or more processor sockets. The processors have additional infrastructure to support extra, built-in processor busses to support the demand of larger systems.

1.3.

The server provides high bandwidth I/O interfaces for connecting to external I/O expansion devices or remote I/O without reducing the number of processor sockets that can be connected together. These may be proprietary interfaces or standard interfaces such as PCIe. The high performance I/O controller to support these slots may be embedded within the main processor socket or on the system board.

2.

Memory RAS and Scalability — All of the following capabilities and characteristics shall be present:

(a)

Provides memory fault detection and recovery through Extended ECC;

(b)

In x4 DIMMs, recovery from failure of two adjacent chips in the same rank;

(c)

Memory migration: Failing memory can be proactively de-allocated and data migrated to available memory. This can be implemented at the granularity of DIMMs or logical memory blocks. Alternatively, memory can also be mirrored;

(d)

Uses memory buffers for connection of higher speed processor -memory links to DIMMs attached to lower speed DDR channels. Memory buffer can be a separate, standalone buffer chip which is integrated on the system board, or integrated on custom-built memory cards. The use of the buffer chip is required for extended DIMM support; they allow larger memory capacity due to support for larger capacity DIMMs, more DIMM slots per memory channel, and higher memory bandwidth per memory channel than direct-attached DIMMs. The memory modules may also be custom-built, with the memory buffers and DRAM chips integrated on the same card;

(e)

Uses resilient links between processors and memory buffers with mechanisms to recover from transient errors on the link; and

(f)

Lane sparing in the processor-memory links. One or more spare lanes are available for lane failover in the event of permanent error.

3.

Power Supply RAS: All PSUs installed or shipped with the server shall be redundant and concurrently maintainable. The redundant and repairable components may also be housed within a single physical power supply, but must be repairable without requiring the system to be powered down. Support must be present to operate the system in degraded mode when power delivery capability is degraded due to failures in the power supplies or input power loss.

4.

Thermal and Cooling RAS: All active cooling components, such as fans or water-based cooling, shall be redundant and concurrently maintainable. The processor complex must have mechanisms to allow it to be throttled under thermal emergencies. Support must be present to operate the system in degraded mode when thermal emergencies are detected in system components.

5.

System Resiliency — no fewer than six of the following characteristics shall be present in the server:

(a)

Support of redundant storage controllers or redundant path to external storage;

(b)

Redundant service processors;

(c)

Redundant dc-dc regulator stages after the power supply outputs;

(d)

The server hardware supports runtime processor de-allocation;

(e)

I/O adapters or hard drives are hot-swappable;

(f)

Provides end to end bus error retry on processor to memory or processor to processor interconnects;

(g)

Supports on-line expansion/retraction of hardware resources without the need for operating system reboot (‘on-demand’ features);

(h)

Processor Socket migration: With hypervisor and/or OS assistance, tasks executing on a processor socket can be migrated to another processor socket without the need for the system to be restarted;

(i)

Memory patrol or background scrubbing is enabled for proactive detection and correction of errors to reduce the likelihood of uncorrectable errors; and

(j)

Internal storage resiliency: Resilient systems have some form of RAID hardware in the base configuration, either through support on the system board or a dedicated slot for a RAID controller card for support of the server’s internal drives.

6.

System Scalability — All of the following shall be present in the server:

(a)

Higher memory capacity: >=8 DDR3 or DDR4 DIMM Ports per socket, with resilient links between the processor socket and memory buffers; and

(b)

Greater I/O expandability: Larger base I/O infrastructure and support a higher number of I/O slots. Provide at least 32 dedicated PCIe Gen 2 lanes or equivalent I/O bandwidth, with at least one x16 slot or other dedicated interface to support external PCIe, proprietary I/O interface or other industry standard I/O interface.

Appendix C

Test Method

1.   Overview

The following test method shall be used for determining compliance with requirements in the ENERGY STAR Product Specification for Computer Servers and when acquiring test data for reporting of Idle State power and Active State power on the ENERGY STAR Power and Performance Data Sheet.

2.   Applicability

The following test method is applicable to all products eligible for qualification under the ENERGY STAR Product Specification for Computer Servers.

3.   Definitions

Unless otherwise specified, all terms used in this document are consistent with the definitions contained in the ENERGY STAR Product Specification for Computer Servers.

4.   Test setup

4.1.

Input Power: Input power shall be as specified in Tables 6 and 7. The frequency for input power shall be as specified in Table 8.

Table 6

Input Power Requirements for Products with Nameplate Rated Power Less Than or Equal to 1 500 watts (W)

Product Type

Supply Voltage

Voltage Tolerance

Maximum Total Harmonic Distortion

Servers with alternating current (ac)-direct current (dc) Single-Output Power Supply Units (PSUs)

230 volts (V) ac or 115 V ac (11)

+/– 1,0 %

2,0 %

Servers with ac-dc Multi-Output PSUs

230 V ac or 115 V ac (11)

Optional Testing Conditions For ac-dc (Japanese Market)

100 V ac

Three-phase Servers (North American Market)

208 V ac

Three-phase Servers (Europe Market)

400 V ac


Table 7

Input Power Requirements for Products with Nameplate Rated Power Greater Than 1 500 W

Product Type

Supply Voltage

Voltage Tolerance

Maximum Total Harmonic Distortion

Servers with ac-dc Single-Output PSUs

230 V ac or 115 V ac (11)

+/– 4,0 %

5,0 %

Servers with ac-dc Multi-Output PSUs

230 V ac or 115 V ac (11)

Optional Testing Conditions For ac-dc (Japanese Market)

100 V ac

Three-phase Servers

(North American Market)

208 V ac

Three-phase Servers (Europe Market)

400 V ac

Table 8

Input Frequency Requirements for All Products

Supply Voltage

Frequency

Frequency Tolerance

100 V ac

50 hertz (Hz) or 60 Hz

± 1,0 %

115 V ac

60 Hz

230 V ac

50 Hz or 60 Hz

Three-phase

(North American Market)

60 Hz

Three-phase

(Europe Market)

50 Hz

4.2.

Ambient Temperature: Ambient temperature shall be within 25 ± 5 °C.

4.3.

Relative Humidity: Relative humidity shall be within 15 % and 80 %.

4.4.

Power Analyser: The power analyser shall report true Root Mean Square (RMS) power and at least two of the following measurement units: voltage, current, and power factor. Power analysers shall possess the following attributes:

(a)

Compliance: The power analyser shall be chosen from the list of power measuring devices specified in the Server Efficiency Rating Tool (SERT)TM  (12) Design Document 1.0.0 (13).

(b)

Calibration: The analyser shall have been calibrated within a year of the test date, by a standard traceable to the National Institute of Science and Technology (USA) or a counterpart national metrology institute in other countries.

(c)

Crest Factor: An available current crest factor of 3 or more at its rated range value. For analysers that do not specify the current crest factor, the analyser must be capable of measuring an amperage spike of at least 3 times the maximum amperage measured during any 1 second sample.

(d)

Minimum Frequency Response: 3,0 kHz.

(e)

Minimum Resolution:

(1)

0,01