Achieving Energy Efficiency in Data Centers

Electrical infrastructure in New Jersey. (Photo: Rich Miller)

In this edition of Voices of the Industry, Brian Korn, Vice President of Data Center Computing for Advanced Energy, explores how switching to 48V can dramatically reduce data center power consumption and meet energy efficiency goals. .

Brian Korn, Vice President of Data Center Computing for Advanced Energy

One of the biggest challenges facing global data centers is the need to meet ever-increasing demands while improving energy efficiency.

Data center expansion continues to intensify, with the global data center market expected to grow in the coming years and energy consumption expected to increase at an annual rate of four to five percent. Although efforts to reduce the energy footprint and improve efficiency have had a significant impact on the computing capacity and energy consumption of data centers, there is no doubt that greater innovation will be necessary as data centers continue to grow in both power consumption and complexity.

If you consider a data center with an energy consumption of 10 MW, the servers can consume 50% of this energy with a power utilization efficiency (PUE) of 1.6. A 2% increase in the energy efficiency of the server power supply results in a 1.6% decrease in electricity consumption. This translates to 1.4 million kWh saved each year, which equates to a reduction in CO2 emissions of over £21 million. At $0.07 per kilowatt hour, that’s $98,000 in savings. In large data centers, which may have even higher power consumption and billing rates, these types of savings add up quickly.

Data centers already account for at least three percent of global electricity consumption and there is intense pressure to increase processing capacity and effectively manage power consumption.

Traditionally, data centers have used 12V power shelves, but as higher performance compute and storage platforms demand more power, moving from 12V to 48V offers inherent efficiency gains, with a reduction in current consumption by a factor of four, which equates to 16 times less distribution losses. This ultimately reduces costs through significantly better thermal performance, optimized efficiency and increased power density.

One of the main economic drivers for data centers to achieve equipment profitability comes from standardization. The adoption of Open Compute standards-based solutions has occurred, in part, due to the need for more efficient data centers. This meets global imperatives for lower carbon emissions, lower power consumption and lower operating costs compared to traditional data centers.

While approximately 15% of data centers have already adopted 48V power architectures, it is estimated that half of leading hyperscalers will move to 48V by the middle of this decade, with traditional enterprise deployments following suit. not as standards are more widely adopted. .

Addressing demands for increased power density and rack payload at the component and system level, Advanced Energy supports the transition from conventional 12V power schemes to 48V architectures with its D-Series. Power Shelves Compliant with Open Compute Project (OCP) ORv3. Developed in conjunction with leading OCP users, the 48V Power Rack aims to create a common power platform for customers across deployments, which should lead to increased adoption and cost savings. scale that benefits all OCP users.

As the industry aims to increase data center efficiency while minimizing environmental impact, cutting-edge technologies such as those developed by Advanced Energy will enable hyperscale and data center customers to efficiently deliver high performance computing infrastructure.

Brian Korn is the Vice President of Data Center Computing at Advanced Energy and brings extensive experience in embedded power solutions for data center computing, hyperscale, telecom and networking products. Contact Advanced Energy to learn more about how their OCP-compliant platform is a catalyst for bringing interoperability, increased reliability of compute and storage applications, and power savings to large-scale deployments, enterprise data centers and edge computing.

Ramon J. Espinoza