Q&A: Mark Bramfitt on Utilities and Data Centers

1 comment

Mark Bramfitt is a leading expert on the relationship between the data center and electric utilities. After years of working with data center operators in a position at California utility PG&E, Bramfitt is now consulting on energy efficiency and power management. Last week on his blog Mark shared some thoughts on the recent Facebook-Greenpeace controversy over power sourcing for data centers.

He wrote: “The resolution of this conundrum, in my view, is an increased partnership between the industry and utilities, addressing energy efficiency, demand response, and yes, power sourcing. The IT industry as a whole needs to engage proactively in these matters to avoid potentially damaging criticism now and in the future.”

How does this come about? What’s the path forward? Mark graciously agreed to an email question-and-answer session to share some of his insights.

DCK: You’ve cited the need for increased partnership between the data center industry and utilities. What’s happening on that front right now?

Bramfitt: I see some good engagement, especially from developers and operators of what I call “utility scale” data centers, on energy efficiency. New multi-megawatt facilities have far better efficiency performance than just a few years ago, and some of these large facilities are being retrofitted to improve performance.

There are two imperatives that now face the industry: moving the energy efficiency needle in the vast market of mid- and small-sized data centers, and addressing more opportunities for engagement and partnerships with utilities.

I envision the second pathway leading to cooperation on demand response and load shifting, as well as addressing the environmental impacts of electric use. The utility-scale users will again take the lead, and they have tremendous market power and profile, but that level of partnership hasn’t begun to take shape in my experience.

DCK: What are some short-term steps that might help the two industries work together more effectively?

Bramfitt: Well, I don’t want to spout the typical platitudes about establishing better lines of communication, etc. But let me step out and say this: of the 12, or 25 (but I don’t think it’s 50) leading utility-scale data center operators, some are going to see the strategic value of pursuing a higher level of engagement regarding electric supply and use.

These breakout firms are going have the discussions with utility partners on how to manage a typical 40 megawatt load. They’re going to do some load shifting with thermal energy storage, backup the grid with their emergency generation capacity, and work with the utility to source clean or renewable power.

DCK: What do you see as the best way for data centers to boost their use of renewable energy? Can utilities address these issues in key markets, or must large data center providers develop on-site generation of renewables?

Bramfitt: It’s a real mixed bag from a regulatory standpoint across the country. In some cases, a customer can directly source generation, and therefore they have the opportunity to pick their source. In other places the utility has a renewable portfolio requirement, so you might get ten percent renewable now and twenty percent in 2020.

So, utility-scale operators really have two main avenues for boosting the use of cleaner power in their facilities: picking the right location (and therefore utility) in the first place, and acting as an advocate in the regulatory arena for changes in the utility industry.

DCK: How might the focus on renewable power – and the prospect of carbon regulation – impact the geography of the data center industry and where new facilities are built?

Bramfitt: I’ve taken a look at this, and believe it or not, the answer is “less than you might think”.

First, what carbon regulation? I don’t see the US charting a carbon regulation scheme in the next few years, or putting one into action in five years. And in the absence of any sense of certainty, businesses simply won’t base their decisions on the “specter” of future taxes.

Many states have set a course for increasing amounts of renewable power in utility generation portfolios – a great thing, but it fails to send a clear price signal because utilities haven’t been able to figure out how to meet the requirements, let alone how expensive it will be.

But let’s say we add a straightforward carbon tax next year born by utilities and therefore their ratepayers. If you are getting 100% coal-fired generation at a retail rate of five cents a kilowatt-hour, I might see a staggering percentage raise of say fifty percent.

If you are sourcing your power from a utility with a much cleaner generation portfolio – natural gas, nuclear, hydro, and some renewables, the percent increase would be much lower. But your existing rates are already much higher – usually twelve cents or more per kWh.

In my estimation then, power price remains the primary consideration for a utility-scale data center operator, and carbon taxation, while raising rates generally, won’t change the relative price disparities that are driving site choices today.

About the Author

Rich Miller is the founder and editor-in-chief of Data Center Knowledge, and has been reporting on the data center sector since 2000. He has tracked the growing impact of high-density computing on the power and cooling of data centers, and the resulting push for improved energy efficiency in these facilities.

Add Your Comments

  • (will not be published)

One Comment

  1. I have followed this space - as it relates specifically to data centers - since the electricity markets deregulated. Having a Father in Law who was Chairman of the American Public Power Association also helped understand the way things work... I look at this a different way completely and think about these questions: How many megawatts of electricity are needed to service a doubling of data center load? Look at analysts projections and put them up against construction of new or upgraded power plants. Do they match? How many megawatts of electricity are being constructed today by utilities (public and private)? How many are planned for the next 20 years? Can the grid accept and effectively manage that load reliably? Utilities I talk with won't look at projects under 200-300 megawatts - it isn't worth the hassle of financing, bonds, permitting etc.to build anything smaller. They also won't put a shovel in dirt until they have commitment as to the load the data center will consume. Of course there are exceptions, but I hear one a year if I am lucky. Data centers - coupled with market expectations around reserving power vs. drawing power - skew things. Let's say a data center has 20 critical megawatts provisioned for their facility. They have sold/leased space drawing 10 megawatts, in reality they are drawing 5, and go to the utility and say 'we need more power to expand' and the utility says, 'you are only using 5 - if you pay for the additional power, you can have it'. One path forward that I am exploring is building my own low carbon power plant on site with my data center buildings. I am self sufficient, can push excess electrons - especially as the first customers move in and I have them drawing 1 MW from a 50 or 100 MW power source - onto the grid, back it up with generators running biodiesel, and I solve a lot of problems. It's not cheap to do this, however what are the valid alternatives? I would love to know where the attractive sites are for data center owner operators where there is abundant cheap power that would also make a great site for a data center. Ashburn grew up around the network density, same with San Jose/Santa Clara/Bay Area not power or the cost of it. In so far as the carbon tax is concerned, there will be one - Congress has discussed it and taxes = revenue in Government. We have never had less tax on anything in my recollection in the U.S. Anyhow, there will be a carbon tax, and whether or not it is enough to offset the premium of 'green' electricity is only looking at one side of the equation - the carbon credits the utility receives which won't be passed onto rate payers are a new form of currency. They can sell those credits to the highest bidder (and polluters) and cha-ching for the utility. then what will happen is they will get pressure from the top 10 ratepayers to share in the carbon credit flow and the rates come down for utilities that have figured it out. The big question I have is will public power or investor owned utilities figure it out first...