Winston Saunders has worked at Intel for nearly two decades and currently leads server and data center efficiency initiatives. Winston is a graduate of UC Berkeley and the University of Washington. You can find him online at “Winston on Energy” on Twitter.
My two days earlier this March at the 2013 Green Grid Forum in Santa Clara were well spent. There were many highlights. To name a few: The roll-out of an equipment disposal metric. earthquake and natural-disaster-proof data center design in Japan and an informative panel discussion on the efficiency challenges of Exascalar with leading industry experts.
But the real game changer for me was the presentation by Dean Nelson (Vice President, Global Foundation Services, eBay) of eBay’s Digitial Service Efficiency indicator, which is available online here: dse.eBay.com (See DCK coverage of the announcement eBay's DSE: One Dashboard to Rule Them All.)
The eBay team has solved two of the most persistent problems in IT -- with full disclosure on how to do it.
The first persistent problem is getting the IT, Finance, Business Unit, and Facilities teams on the same page. Now to be fair, eBay is not the only company doing this. Joe Kava (Vice President - Data Centers at Google) has also disclosed Google's organizational structure to optimize the overall efficiency of Google's most costly asset (i.e. its data centers). But where eBay breaks the mold is by disclosing the next level of detail. As Stanford professor Jon Koomey says so eloquently: “Fixing misplaced incentives is the most important step toward realizing (the potential for efficiency improvement).”
The second problem eBay solved is communicating the detailed management of its core business, as shown in the Digital Service Efficiency indicator reproduced above.
An important element of this data is that through open disclosure, it permits us to quantify the sustainability benefits of our digital economy. It’s relatively easy to calculate the carbon and water impact of each transaction from the data. For instance, for the roughly 46,000 transactions/kWh recorded in 2012, eBay used about 0.053 mL of water (See calculation) and about 22 mg (See calculation) of carbon. To put this into perspective, this equates to about one drop of water (See calculation) and about a mass of CO2 equivalent to a small grain of rice (See calculation) per each transaction. .
That’s impressive when you consider the carbon footprint of a physical transaction. Imagine I drive to a store five miles from my home to make a similar purchase. Assuming an efficiency of 20 mpg, I’ll end up generating about 4.5 kg of CO2 (See EPA calculation) for the round trip, or about the mass of a decently sized kettlebell. Here I’ve associated driving with the transaction only. One might argue shipping and other factors should be included, but those details vary greatly, and really only obscure the comparison of virtual to physical transaction cost. The main point is that doing transactions through a well-managed data center has an astounding reduction in resource impact compared to physical activity: in this case -- almost a factor of 200,000.
True Game Changer
That’s why eBay has changed the game. They have provided a role model, which is organized around common metric, to optimize the overall effectiveness of IT, and have openly disclosed the metrics and indicators they use to do it. This has the huge side benefit that we, as eBay customers, can understand the impacts of our own actions in terms of carbon and water use. And while there is at this point scant publicly available data on water use, eBay’s early leadership using the Green Grid’s WUE metric sets the stage for greater openness throughout the industry.
Kudos to the eBay team for their openness, this disclosure breakthrough, and their leadership in data center management.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.