Next-Generation Data Center Facility White Paper
Huawei Digital Power
Developed by Huawei with industry experts, the paper defines a next-generation facility by four attributes: sustainable, simplified, autonomous-driving (automated O&M) and reliable. It argues for moving from PUE alone to a family of xUE metrics.
On this page
Overview
According to the paper, PUE of global large data centres stayed around 1.6 for seven years (Uptime 2021), while Chinese policy targets new large data centres below 1.3 by 2025. The authors therefore advise looking beyond PUE to CUE, WUE and other xUE metrics, heat reuse and recyclability.
On speed and space, prefabricated modular build can cut a 1000-rack data centre from more than 18 months to 6-9 months. In one cited project the PowerPOD reference saves more than 40 % of power-system space, allowing 350 more IT racks and more than 16,000 m less power cable. Above about 20 kW per rack, liquid cooling is expected to become the primary method with air as auxiliary (converged air-liquid cooling).
The operations and reliability section rests on automation. A 1000-rack data centre typically needs 15-30 O&M engineers doing 6-12 on-site inspections, while infrared and robotic inspection covers 2000 racks remotely in 5 minutes. AI energy optimisation at Huawei Cloud Langfang (700+ points) cut annual PUE from 1.42 to 1.25, saving 33.36 GWh. Per Uptime 2021, power and cooling failures account for 57 % of outages (43 % power, 14 % cooling).
Key points
- PUE of large DCs stayed around 1.6 for seven years (Uptime 2021); Chinese policy targets new large DCs below 1.3 by 2025.
- Prefabricated modular build cuts a 1000-rack DC from more than 18 months to 6-9 months.
- PowerPOD reference saved >40 % of power-system space: 350 more IT racks and >16,000 m less power cable.
- Above about 20 kW per rack liquid cooling becomes primary, with air as auxiliary.
- A 1000-rack DC needs 15-30 O&M engineers and 6-12 on-site inspections; robotic inspection covers 2000 racks in 5 minutes.
- Huawei Cloud Langfang AI optimisation cut annual PUE from 1.42 to 1.25, saving 33.36 GWh.
- Uptime 2021: 57 % of outages come from power and cooling (43 % power, 14 % cooling).
Using it in design
- Support CUE, WUE and a heat-reuse fraction next to PUE in reports.
- In the cooling advisor, suggest liquid cooling above about 20 kW per rack.
- In ops planning, estimate O&M headcount from rack count and automation level.
- In reliability checks, weight power-chain design heavily, since power causes the largest outage share.
Monthly newsletter · free
Keep your reading list current
Once a month: new library sources, changes to Turkish regulations and upcoming data centre events.
SoSo CAD
Compare the cost of design decisions
Evaluate equipment costs, energy expenses and total cost of ownership (TCO) using your inputs and assumptions in SoSo CAD.
Request a demo
Who should read it
Facility investors and designers planning metrics, operations and fast deployment together.
Related standards
- GB 50174-2017
- The Green Grid xUE metrics
Publication details
- Authors
- Huawei Digital Power
- Publisher
- Huawei Digital Power
- Year
- 2022
- Document language
- English
- Source type
- Manufacturer
- Access
- Free to read
- Licence
- Copyright retained; citation only — the publisher’s terms apply to the original document
Citation
Huawei Digital Power (2022). Next-Generation Data Center Facility White Paper. Huawei Digital Power. https://digitalpower.huawei.com/attachments/index/1c651408309b41c78e368a270f89d89e.pdf
Events on this subject
Browse the events calendarAll content and rights belong to the respective authors and publishers. SoSo CAD provides only the citation, its own summary and a link to the source, and is not responsible for the sources' content.
Related sources
All sources on this topicHuawei Digital Power 2023
White Paper on the Top 10 Trends of Data Center Facilities
Show WUE alongside PUE in the metrics panel, with region-specific targets as policy presets.
- At 16 kW per rack, power-room area roughly equals IT space (1:1).
arXiv preprint (cs.AI), Harvard T.H. Chan School of Public Health 2026
Assessing the Carbon Emissions and Energy Consumption of U.S. Hyperscale Data Centers
In fleet or facility energy estimation, use capacity × 8,760 × facility-load coefficient (0.48-0.70, central 0.58) and report the range (67.7-98.6 TWh) explicitly.
- State-level intensities (central scenario): Virginia 535, Oregon 440, Iowa 645, Nebraska 740, Wyoming 1035 gCO2/kWh; Virginia had 142 HDCs, 21.0 TWh, 11.5 Mt.
arXiv preprint (math.OC), MIT Operations Research Center 2025
Locational Marginal Emissions for Carbon-Aware Data Center Operations in Large-Scale Power Grids
In data-centre carbon accounting, consider hourly locational marginal emissions (LME) rather than an annual-average grid factor; the LME-based intensity (about 672 kgCO2/MWh) was about 3 times the average factor (202).
- Emission rates assumed in the simulated grid: coal 800 kgCO2/MWh, gas 450 kgCO2/MWh, wind/solar/nuclear/hydro 0.
arXiv preprint (cs.SE), Google 2024
Carbon accounting in the Cloud: a methodology for allocating emissions across data center users
In allocating carbon to data-centre users, give idle power to whoever holds the resource allocation and share dynamic power by actual usage; because an idle server draws about 45% of peak power, the idle share is large.
- Resources with equivalent busy power used to weight idle-power allocation: 1 Intel Skylake N1 CPU core = 20 GiB RAM = 1 TiB SSD = 6 TiB HDD.
European Commission (Official Journal of the European Union) 2024
Commission Delegated Regulation (EU) 2024/1364 of 14 March 2024 on the first phase of the establishment of a common Union rating scheme for data centres
Build the metering architecture to the regulation: meters on every supply input for total energy (grid and back-up generator separately) and at every UPS output for IT energy, with a system that retains measurement point and device records for at least 10 years.
- Data centre electricity demand is projected to reach 3.2% of total EU demand by 2030, a 28% increase over 2018.
Proceedings of the Twelfth ACM International Conference on Future Energy Systems (ACM) 2021
A Guide to Reducing Carbon Emissions through Data Center Geographical Load Shifting
For geographic load shifting, consider locational marginal carbon emissions (lambda-CO2) as the signal instead of LMP or regional average carbon intensity; it gave the largest reduction (-1.44%).
- Lambda-CO2 shifting saved about 0.14 t CO2 per MW shifted and still reduced generation cost by 0.27%; LMP shifting cut cost by 1.07% but saved less carbon.