I write today’s blog knowing that most of my colleagues at BNP Media have already decamped for a long weekend and also knowing that I’ll be spending this very important holiday touring a data center in Switzerland, where Memorial Day is not a big deal, I’m told. Ironically, I just had a French executive wish me a nice long weekend. He apparently will be here on Monday, enjoying a barbecue, along with everyone else.
Well, c'est la vie!
Just a few moments ago, I was on the phone with Jean-Simon Venne, vice president of Energy Efficiency for SM Group International. He made an interesting case that SM Group had developed an innovative way to manage airflow to reduce energy use in data center facilities. As it was an introductory conversation, I don’t really know what’s in SM Group’s secret sauce. Venne did begin his overview by talking about thermodynamics. He ended by saying, “Sometimes it feels more like chemistry than engineering.” I’ll be following up.
A few minutes after this conversation, I realized that the message about PUE from designers, builders, and vendors has shifted. The emphasis that was once placed on achieving ultra-low PUEs is now on achieving ultra-low PUEs at comparatively low prices.
I suppose this shift was inevitable, as more and more companies wanted to brag about achieving the same PUEs available to internet companies. At first, their achievements were dismissed because these businesses are unique and their data centers do not resemble enterprise-class data centers. Now, though, businesses are beginning to develop solutions that they claim will enable enterprises to achieve ultralow PUEs, at least in new data centers.
Customers, I know, will begin asking these providers for more details about how ultralow PUEs can be achieved in heterogenous server environments. If the solutions withstand scrutiny, it will certainly vindicate those who thought that injecting a little competition into the energy efficiency conversation would pay off.