On a recent flight, I had the pleasure of watching a number of episodes of E2, a PBS show narrated by Morgan Freeman. The shows cover a number of environmental topics and features Stephen Chu, former energy expert at LBL and now Secretary of Energy. If you haven't seen them, I highly recommend finding a few episodes. Unfortunately, only the "design" episodes are on hulu, the "energy" and "transport" episodes have not been posted online.
A couple of things stood out at me, and I came away with a feeling that we could do better at achieving energy efficiency and energy independence. Not just because we should as a planet-saving activity, but because we've actually gone backwards in efficiency in the last 100 years.
First, consider this statistic: The first Ford Model T accepted both Ethanol and Gasoline, and got 25 miles to the gallon. Today's cars average 21 miles to the gallon, and while it only costs $50 to outfit a gasoline car at production with Ethanol capabilities, few cars are actually "FlexFuel".
Now, you may wonder what this has to do with data centers, but the answer is "a lot!" 74% of data centers refuse to enable power saving functionality if it affects performance (according to a 2008 study by Aperture Research Institute). The same is true with cars -- when we can increase efficiency, consumers tend to ask for more performance or torque at the same cost (rather than a more efficient vehicle). Given that computing cycles are a commodity, it is astounding that distributing applications more intelligent is not on the radar of IT managers.
At Sentilla, we have a customer that is a large manufacturer. They factor in the 3-year cost of energy in all of their CapEx purchases. They've also pushed the operating energy budget to the part of the organization that owns the assets. In the case of trucking, there was an overnight change in behavior. Buying "big, awesome trucks" with the lowest capital cost was no longer the norm; all of a sudden, energy efficient trucks were acquired at higher capital cost due to the fact that their cost over a 3-year period was less. So then why do IT managers rate "energy efficient equipment" as one of the lowest priorities? Maybe because only 24% of companies chargeback energy consumption to those using IT resources. Frankly, based on my experience, 24% is really high and I'd put it at somewhere closer to 5%. Even uber-jumbo companies with more than 1,000,000 square feet of data center space don't charge back the energy cost, only the fractional CapEx of the asset being used.
What astounded me is that 37% of data centers have NO plans to measure energy consumption or consider energy efficiency when purchasing data center equipment -- that's like refusing to look at the EPA fuel mileage when buying a vehicle.
Comments
Model T got 25mpg, really?
regarding: "The first Ford Model T accepted both Ethanol and Gasoline, and got 25 miles to the gallon. Today's cars average 21 miles to the gallon" F = M*A, what did the Model T weigh? and at what speed did it go? Wind drag? Modern fuel mileage calculations are based on an average of driving, a lot of it much, much faster than they went for a Model T (and their were no highways). OR are you telling me that someone has benchmarked a Model T in good tune to get the same equivalent numbers? Then lets talk about all of the modern comfort items on a car that take power (gas): power steering, air conditioning, all of the electrical features, automatic transmissions, wider/safer tires. Those were not found on a Model T. Last year I bought, as a backup car, a 2009 Chevy Cobalt EFI (it was cheap). It has a 5 speed manual transmission, it's light weight, it has low rolling resistance tires (notice how small the footprint is on the Model T tires). I drive it hard and easily get 30mpg with mixed driving. It's rated up to 37mpg highway, I saw 35mpg before the engine was broken in. So there.
Post new comment