The claim to energy efficiency is easily made. It is surprising that quite a few solutions have the statement of being energy efficient. There are a few metrics for energy efficiency around. The most famous in the datacentre industry is the PUE (Power Usage Effectiveness) ratio. The PUE also plays an important role in the European Code of Conduct for datacentres 5 . The challenge is that claimed metrics are hard to get verify.
Who is going to check a 1.15 PUE of a datacentre?
Of course we live in a very honest world where nobody would even think of publicly stating a slightly better metric for commercial reasons. But it is a real challenge to define usable, verifiable and meaningful metrics, even for professional and specialized organizations whose intentions are good. Let alone the preferred idea that metrics are measured and deliver the same outcomes, independent from who is measuring for a certain metric/energy efficiency ratio. The industry, users and buyers are entitled to know the real value of an energy efficiency metric. The claim of an energy efficiency metric needs to be realistic, trustworthy (measurable, verifiable) and aligned with other energy efficiency initiatives, aligned with current standards and methods that are academically sound.
OPERA will conduct research to find the right energy efficiency metrics, derived from the researched model and translated in algorithms that are the best fit for the circumstances, and describing the right way for measuring, comparing and monitoring these metrics.
As reported in the “STRATEGIC RESEARCH AGENDA of EPoSS – THE EUROPEAN TECHNOLOGY PLATFORM ON SMART SYSTEMS INTEGRATION” 4 , energy sources beyond primary batteries should be also investigated. As an example, wind turbines and/or photovoltaic panels may be included as a part of the power supply system. This approach could also lead to a further extension of the operational time of the datacentre in case of the electrical power is provided by local rechargeable batteries.
