This audio was created using Microsoft Azure Speech Services.
“Many data centers use much more energy that they need to,” says Soeren Brogaard Jensen, VP Enterprise Software at Schneider Electric. But historically, data center managers have tended to be charged with guarding the availability of their facilities rather than their energy efficiency or carbon emissions. But as the number of servers being hosted in data centers has accelerated and the cost of energy increased dramatically, efficiency has become a much bigger factor.
Recent studies have shown that energy use is the most substantial cost of IT operations; in the data center, the cost of powering servers can often exceed the cost of purchasing the equipment in the first place. “There’s a significant opportunity to produce energy savings of at least 20% just by implementing a rudimentary data center energy management program,” continues Soeren Brogaard Jensen.
But how simple can an energy management process be, and how few measurements are really needed, in order to provide the information necessary to successfully manage infrastructure energy use and allocate energy costs and carbon to IT users? There is an extremely “low-cost or no cost” solution available right now.
The problem with a lot of energy management programs is that the cost of instrumentation for each data center system can seriously damage ROI to the point where any benefit is completely eroded. To counter this, Soeren Brogaard Jensen describes an extremely simple process with very few measurements, which anyone can implement immediately, and which will provide accuracy that is “good enough” for an effective energy management program.
According to Jensen, there are generally three goals for an assessment of energy efficiency; a one-time benchmarking of performance; to pass-through allocation of energy use or carbon production to data center customers; or to produce data in order to reduce data center energy use and carbon impact. It is important to recognize which of these goals (or combination of goals) is intended, because a correct understanding dramatically affects the technical implementation.
For an overview of energy cost and carbon allocation strategies and their precision, please download APC by Schneider Electric white paper #161; “Allocating Data Center Energy Costs and Carbon to IT Users”. Authored by Neil Rasmussen, the white paper shows that it is both easy and inexpensive for any data center, large or small, new or old, to get started allocating costs and carbon, but the expense and complexity escalate and ROI will decline if excessive precision is specified.
And to learn more about the “Good Enough” energy management process and how to implement it in your data center, listen to our entire 10-minute podcast with Soeren Brogaard Jensen.