Tech Insights

Looking toward the Second Decade of Digital Power Technology

May 01, 2015 by Jeff Shepard

Digital power technology has entered its second decade. Last year’s Darnell Power Forum (DPF ’14) marked the tenth-anniversary of that event. The first

Digital power technology has entered its second decade. Last year’s Darnell Power Forum (DPF ’14) marked the tenth-anniversary of that event. The first DPF a decade earlier kicked-off the commercial development of digital power and included the first meeting of what would become the PMBus organization. As the world’s leading analysts in the development of digital power technology, Darnell Group projects that the emerging area of software-defined power architectures (SDPAs) will be the defining aspect of the second decade of this important area of power electronics technology. One of several themes for DPF ’15 will be; “Dawn of the Second Decade of Digital Power.” 

Other themes at DPG ‘15 will include; emerging semiconductor materials such as GaN, SiC and GaAs; implementing smart grids, micro grids, nano grids and related networking technologies; DC power distribution architectures; and advanced power conversion, energy storage and energy harvesting technologies.

Looking at anticipated trends in digital power, Darnell has developed this definition for SDPAs: “Software-defined power architectures (SDPAs) are systems where elements whose performance has previously been optimized on a local and relatively isolated basis are instead globally and interactively optimized in real-time through software control to achieve improvements in operational efficiency, availability, reliability, and cost. SDPAs include the optimization of power distribu-tion, power conversion, and power consumption on both a global and granular basis.”

Successful development of SDPAs will demand technology advancements in all areas of power electronics, including the deployment of new semiconductor materials such as GaN and GaAs to enable advanced power conversion architectures as the “foundation” upon which complex and comprehensive SDPAs will be built.

SDPAs will be required to support 100GB optical networking, and are seen to be absolutely necessary to the successful launch of next-generation 5G wireless technologies. There are current (and early on) development activities at Huawei, Ericsson, Bell Labs, Intel, Google, the ITU, at various universities around the world, and in many other organizations we have yet to identify. Developments in Smart Grid technologies, Demand Management programs, Real-time Pricing deployments, and so on, are also driving interest in SDPAs.

“The potential of software-defined power can be reached if the industry reaches a consensus on a solution with reference architecture and common standards. This will help to provide power to data centers based on demand consumption, rather than planning and provisioning the power and cooling requirements based on preexisting knowledge of peak systems usage,” according to a white paper issued by Cognizant.

This is a broad and expansive area of technology. No single company is expected to dominate all aspects of SDPAs. These implementations will extend from “in package” power conversion based on new semiconductor materials to high-level “energy aware” designs stretching across the world and using real-time dynamic energy management schemes yet to be developed. In addition, we see continued advancements in metrology-related technologies for power management applications. Plus the emergence of new semiconductor materials such as GaN and GaAs is already enabling higher-frequency and more compact converters and in next-generation designs these new materials are expected to contribute to the continued development of “in package” power conversion for finer granularity of power management for FPGAs, microprocessors, etc.

SDPAs are being enabled by continued advancement in digital electronics, as well as by advancements in A/D and D/A conversion, power conversion and power management technologies. The concept will continue to evolve and expand as enabled by future generations of the related technologies. There are already several trends/developments that at least “hint” at the emergence of SDPAs. One example is:

A virtual energy consumption meter capable of estimating in real–time the power consumption of networking devices by jointly exploiting easily observable network statistics and empirical power consumption models has been developed and demonstrated by a team of European engineers. The resulting “virtual power meter” gathers the network statistics (such as browsing, video streaming, and background FTP traffic) and estimates the power consumed by the wireless access point in a given observation period. The median values of the actual and estimated power consumption in a demonstration across the 10 runs were, respectively, 3.437W and 3.446W. The median absolute error was 28mW. A network slowdown was generated between the 5th and the 15th seconds of the test, and the virtual power meter closely tracked the actual power consumption during the slowdown.

Europe’s “Smart Energy Aware Systems” (SEAS) program is another example of SDPA-related activities. According to the SEAS web site: Environmental, economic and sustainability challenges of continuously increasing energy consumption are present all over the world. Meeting the challenges requires cross-industry cooperation and the means for consumers to influence their energy consumption in terms of the quantity and type of energy consumed. The SEAS project will address the problem of inefficient and unsustainable energy consumption, which is due to a lack of sufficient means to control, monitor, estimate and adapt the energy use of systems versus the dynamic use situations and circumstances influencing the energy use. 

The objective of the SEAS project is to enable energy, ICT and automation systems to collaborate at consumption sites, and to introduce dynamic and refined ICT-based solutions to control, monitor and estimate energy consumption. An additional aim is to explore business models and solutions that will enable energy market participants to incorporate micro-grid environments and active customers.

In a recent announcement, the IEEE Journal on Selected Areas in Communications “ called this next-generation of digital power, “One of the most urgent and critical challenges in the design of 5G wireless communication systems is that of mitigating their energy consumption in light of a greatly increased data rate. Indeed while it has now been some years that the topic of energy-efficient design of communication networks has gained a prominent role, due to both economic reasons and environmental concerns on sustainable growth, it is expected that in the 5G era, with millions more base stations and billions of connected devices.” As a result, the topic of “software-defined power architectures” (SDPAs) will be even more compelling.

“Given the required 1000x increase in offered data rates and throughput with respect to current state-of-the art technology, the network energy efficiency must be improved by at least the same factor, in order to keep the energy consumption at today’s level. This special issue [to be published in the first-half of 2016] will address research advances that are instrumental in a 1000x capacity improve-ment with respect to currently deployed wireless networks, while achieving at the same time increased energy efficiency levels,” the IEEE announcement continued.

Ericsson announced new collaborations with prestigious universities King’s College London and Technische Universität Dresden (TU Dresden). The collective efforts will focus on 5G research, addressing both the technical implications and the societal challenges of implementing the next-generation of communications technology. The collaborations with the two universities will build on other leading European research institute and university collaborations in the 5G sphere, such as those with the Royal Institute of Technology, Chalmers University of Technology and Lund University in Sweden.

“The Software-Defined Power Architecture is being seen in the industry as the best way forward to optimize energy utilization in data networks and make them more energy efficient. The optimization of energy down to a granular level was at the origin of Ericsson’s digital power research, which was started in 2004,” stated the announcement.

I expect that there are other developments that we have not yet identified that will contribute to the emergence of SDPAs. As you can see, several of these developments are either very recent or are still pending. As a result, I believe that we are only seeing a glimpse of what will ultimately become SDPAs.

In addition to our usual comprehensive coverage at this fall’s Darnell Power Form (DPF ’15), we will be looking forward to the “Second Decade of Digital Power.”  One of the areas we are actively investigating includes developments related to the emergence of SDPAs. As I’ve point out, we believe that the emergence of SDPAs will be the defining development in the Second Decade of Digital Power. We have a broad interest in the area of SDPAs and I want to ask you for your comments and insights into this area. Please email me at [email protected] or call me at +1-951-279-6684. Thank you in advance. I look forward to hearing from you and to the possibility to seeing you at DPF’15 and the Dawn of the Second Decade of Digital Power.