I have written many times on the importance of the site selection for data centers and its growing importance when one considers the regulatory and legislative efforts underway globally. Those who make their living in this space know that this is going to have a significant impact on the future landscape of these electronic bit factories. The on-going long term operational costs over the life of the facility, their use of natural resources (such as power) and what they house and protect (PII data or Personally Identifiable Information) are even now significantly impacting this process for many large global firms, and is making its way into the real estate community. This is requiring a series of crash courses in information security, power regulation and rate structures, and other complex issues for many in the Real Estate community.
In speaking to a bunch of friends in the Real Estate side of the business, I thought it might be interesting to take a few of these standard criteria head on in an open discussion. For this post I think I will take on two of the elemental ones in data center site selection. We will look at one major item, and one item that is currently considered a minor factor that is quickly on the rise in terms of its overall importance. Namely Power and Water, respectively.
Watts the Big Deal?
Many think that power cost alone is the primary driver for data centers, and while it is always a factor there are many other facets that come into play underneath that broader category of Power. While Site Selection Factors are always considered highly confidential I thought it might highlight some of the wider arcs in this category.
One such category getting quite a bit of attention is Power Generation Mix. The Generation mix is important because it is essentially the energy sources responsible how that area or region gets its energy. Despite what politicians would lead you to believe, once an electron is on the grid it is impossible to tell from which source it came. So â€˜Green Energyâ€™ and its multitude of definitions is primarily determined by the mix of energy sources for a given region. A windmill for example does not generate an electron with a tiny label saying that is sourced from â€˜greenâ€™ or â€˜renewableâ€™ sources. Understanding the generation mix of your power will allow you to forecast and predict potential Carbon output as a result of Data Center Carbon production. The Environmental Protection Agency in the US, produces a metric called the Carbon Emission Factor which can be applied to your consumption to assist you in calculating your carbon output and is based upon the generation mix of the areas you are looking to site select in. Whether you are leasing or building your own facility you will likely find yourself falling into a mandatory compliance in terms of reporting for this kind of thing.
So you might be thinking, â€˜Great, I just need to find the areas that have cheap power and a good Carbon Emission Factor right?â€™ The answer is no. Many Site Selection processes that I see emerging in the generic space start and stop right at this line. I would however advocate that one takes the next logical step which is to look at the relationship of these factors together and over a long period of time.
Generation Mix has long been considered to be a â€˜Foreverâ€™ kind of thing. The generation sources within a region, rarely changed, or have rarely changed over time. But that is of course changing significantly in the new era that we live in.
Lets take the interplay (both historical and moving forward) of the Power Cost and its relationship with the Generation Mix. As humans we like to think in simplistic terms. Power costs for a specific region are â€˜so many cents per kilowatt hourâ€™ this changes based upon whether you are measured at a residential, commercial, or industrial rate schedule. The rate schedule is a function of how much power you ultimately consume or promise to consume to the local utility. The reality of course is much more complicated than that. Power rates fluctuate constantly based upon the overall mix. Natural disasters, regulation, etc. can have a significant impact on power cost over time. Therefore its generally wise to look at the Generation Mix Price Volatility through the longer term lens of history and see how a regionâ€™s power costs oscillate between these types of events. However you decide to capture or benchmark this it is a factor that should be considered.
This is especially true when you take this Volatility factor and apply it the changing requirements of Carbon Reporting and impacts. While the United States is unlikely to have a law similar to the CRC in the UK (Carbon Reduction Commitment), it will see legislation and regulation impacting the energy producers.
You might be asking yourself, â€˜Who cares if they go after those big bad energy companies and force them to put more â€˜green power in their mixesâ€™. Well lets think about the consequences of these actions to you the user, and why its important to your site selection activity.
As the energy producers are regulated to create a more â€˜greenâ€™ mix into their systems, two things will happen. The first of course is that rates will rise. The energy producers will need to sink large amounts of capital to invest into these technologies, plants, research and development, etc to come to alignment with the legal requirements they are being regulated to. This effect will be uneven as many areas around the globe have quite a disparate mix of energy from region to region. This will also mean that â€˜greenerâ€™ power will likely result in â€˜more expensive powerâ€™. Assessing an area for the potential impacts to these kinds of changes is definitely important in a data center build scenario as you likely have a desire to ensure that your facility has the longest possible life which could span a couple of decades. The second thing which may be a bit harder to guess at, is â€˜which technologyâ€™ is a given region or area likely to pick and its resulting carbon output impact. While I have a definite approach to thinking through such things, this is essentially the beginning of the true secret sauce to site selection expertise and the help you may require if you don’t have an internal group to go through this kind of data and modeling. This is going to have an interesting impact on the â€˜clusteringâ€™ effect that happens in our industry at large.
We have seen many examples like Quincy, Washington and San Antonio, Texas where the site selection process has led to many Data Center providers locating in the same area to benefit from this type of analysis (even if not directly exposed to the criteria). There is a story (that I donâ€™t know if its true or not) that in the early days when a new burger chain was looking to expand where it would place its restaurants, it used the footprint of its main competitor as its guide. The thinking was that they probably had a very scientific method for that selection and they would receive that same ancillary benefit without the cost and effort. Again, not sure if that is true or not, but its definitely something likely to happen in our industry.
In many markets these types of selections are in high demand. Ascent Corporation out of St. Louis is in the process of building a modern facility just down the street from the Microsoft Mega-Facility near Chicago. While Ascent was a part of the original Microsoft effort to build at that location, there has been an uptick in interest for being close
to that facility for the same reasons as I have outlined here. The result is their CH2 facility is literally a stones throw from the Microsoft Behemoth. The reasons? Proximity to power, fiber, and improved water infrastructure are already there in abundance. The facility even boasts container capabilities just like its neighbor. The Elmhurst Electrical Substation sits directly across the highway from the facility with the first set of transmission poles within easy striking distance.
The Generation mix of that area has a large nuclear component which has little to no carbon impact, and generates long term stability in terms of power cost fluctuations. According to Phil Horstmann, President of Ascent, their is tremendous interest in the site and one of the key draws is the proximity of its nearby neighbor. In the words of one potential tenant â€˜Its like the decision to go to IBM in the 80s. Its hard to argue against a location where Microsoft or Google has placed one of its facilities.â€™
This essentially dictates that there will be increasing demand on areas where this analysis is done or has been perceived to be done. This is especially true where co-location and hosting providers can align their interests with those commercial locations where there is market demand. While those that follow first movers will definitely benefit from these decisions (especially those without dedicated facility requirements), first movers continue to have significant advantage if they can get this process correct.
Tying into the power conversation is that of water. With the significant drive for economization (whether water based or air-based) water continues to be a factor. What many people donâ€™t understand is that in many markets the discharge water is clean to dump into the sewage system and to â€˜dirtyâ€™ to discharge to retention ponds. This causes all kinds of potential issues and understanding the underlying water landscape is important. The size of the metropolitan sewage environments, ability to dig your own well efforts, the local water table and aquifer issues, your intended load and resulting water requirements, how the local county, muncipality, or region views discharge in general and which chemicals and in what quantities is important to think about today. However, as the use of water increases in terms of its potential environmental scrutiny â€“ water is quickly rising on the site selection radar of many operators and those with long term holds.
I hope this brief talk was helpful. I hope to post a few other key factors and a general discussion in the near future.