ENERGY STAR Product Specification for Data Center Storage: Test Method (Rev. April 2020)

Overview

The document opens with a note that EPA coordinated with SNIA on the Emerald Version 4.0.0 test method and incorporated it into the testing protocol for the ENERGY STAR Version 2.0 data-centre storage specification. EPA changed no other element of the test method and understands that results from the previous Emerald method may be used to certify to Version 2.0. The method applies to all products eligible under the specification, with terms as defined in the eligibility criteria.

Test setup rests on three conditions: ambient temperature between 18 C and 28 C throughout, relative humidity between 15% and 80%, and input power per Tables 1 and 2. For products rated 1500 W or less, voltage tolerance is +/-1.0% and total harmonic distortion at most 2.0%; above 1500 W the figures are 5.0% and 5.0%. Supply options are 230 V or 115 V AC, with optional 100 V AC for the Japanese market, 208 V three-phase for North America and 400 V three-phase for Europe. Frequency tolerance is 1.0%: 50 or 60 Hz for 100 V, 60 Hz for 115 V, 50 or 60 Hz for 230 V and three-phase.

Instrument requirements are detailed. The power meter must report true RMS power and at least two of voltage, current and power factor, be calibrated within the past year to a NIST-traceable standard, and have an available current crest factor of at least 3 and a frequency response of at least 3.0 kHz. Resolution is 0.01 W below 10 W, 0.1 W for 10-100 W and 1.0 W above 100 W, with at least one measurement set per second. Uncertainty is at most 2% (95% confidence) from 0.5 W. The temperature sensor logs at least 4 samples a minute, measures no more than 50 mm in front of the main airflow inlet and has +/-0.5 C accuracy or better.

Test conduct refers to SNIA Emerald 4.0.0: pre-fill, conditioning, block-access active state, file-access active state, ready idle and capacity-optimization (COM) validation tests. Where a product ships with COMs, those that can be disabled must be off during pre-fill, conditioning, active-state and idle tests, then enabled after the idle test for COM validation. Products certified on block I/O capability take the block I/O active-state test, and those certified on file I/O take the file I/O active-state test. COM verification is run once with storage devices of the vendor's choice and need not be repeated with other devices.

Key points

1. EPA updated the test method to incorporate SNIA Emerald Version 4.0.0; earlier Emerald results may still be used for Version 2.0 certification.
2. Ambient temperature during testing must stay between 18 C and 28 C, relative humidity between 15% and 80%.
3. For products rated 1500 W or less, voltage tolerance is +/-1.0% and THD at most 2.0%; above 1500 W it is 5.0% and 5.0%.
4. Supply options: 230 V or 115 V AC, optional 100 V AC (Japan), 208 V three-phase (North America), 400 V three-phase (Europe); frequency tolerance 1.0%.
5. Power meters must report true RMS, have a current crest factor of at least 3, a frequency response of at least 3.0 kHz, annual calibration and at least one reading per second.
6. Power measurement uncertainty is at most 2% at 95% confidence for 0.5 W and above; resolution is 0.01 W below 10 W and 1.0 W above 100 W.
7. Temperature sensors log at least 4 samples a minute, measure no more than 50 mm in front of the air inlet and have +/-0.5 C accuracy.
8. COMs that can be disabled are turned off for pre-fill, conditioning, active-state and idle tests and enabled afterwards for COM validation.

Using it in design

• Record the test-condition envelope (18-28 C inlet, supply voltage) alongside certified storage power values so users know published watts correspond to a cool inlet; apply a temperature-uplift caveat when the design inlet setpoint is higher.
• Reuse the analyzer criteria (RMS, crest factor 3, 3.0 kHz, 2% uncertainty, annual calibration) in the commissioning module's metering checklist.
• For storage heat and power verification, plan inlet-air temperature measurement no more than 50 mm from the product inlet.

Who should read it

Laboratories running ENERGY STAR storage tests, manufacturers, and design or commissioning engineers interpreting the test conditions.

Document contents

1. Overview
2. Applicability
3. Definitions
4. Test Setup
5. Input Power
6. Ambient Temperature
7. Relative Humidity
8. Power Meter
9. Temperature Sensor
10. Test Conduct
11. Guidance for Implementation of SNIA Emerald Power Efficiency Measurement Specification Version 4.0.0
12. Test Procedures for All Products
13. SUT Pre-fill Test
14. SUT Conditioning Test
15. Block I/O Active State Test
16. File I/O Active State Test
17. Ready Idle State Test
18. COM Validation Test

Related standards

• SNIA Emerald V4.0.0
• ENERGY STAR Data Center Storage V2.1

Read the document

Publication details

Authors

U.S. Environmental Protection Agency

Publisher

U.S. Environmental Protection Agency (ENERGY STAR)

Year

2020

Document number

Rev. Apr-2020

Document language

English

Source type

Government / public

Access

Open licence — downloadable

Licence

U.S. public domain — U.S. Government work; free to reuse

Licence evidence

www.copyright.gov

Citation