Across the United States, data centers consume as much as power as five million homes. A 50,000-square-foot data center - small in comparison to the hyper-scale data centers being utilized for giants such as Amazon, Google, Oracle, and Microsoft - can consume five megawatts of power or enough energy to power 50,000 homes. Google has reported using enough energy to power 200,000 homes, and its data centers around the globe are reported to draw over a combined 260 million watts of power.
Equally important is the fact that companies cannot afford to experience business disruption due to a data center failure. Potential locations with the fewest incidents of natural disasters can move into a very favorable position in the data center site selection process.
One trending method of saving capital investment costs is through the development and use of modular data centers. These mobile units look more like a big cargo container than a data center, but they provide a small capital-intensive start and have the advantage of being able to be moved. Land acquisitions where the mobile units can be placed become more of a focus than building a permanent structure, and the speed and ease of putting these units into action, teamed with their energy efficiency, can reduce capital costs and energy expenditure.
Location Modeling and Quantitative Analysis
The data center site selection process begins with a set of project assumptions from the company. These assumptions include geography, capital investment, facility size, utility requirements, real estate requirements, and telecommunications redundancy, among other factors. Based upon these assumptions, data is gathered through research and information from public and proprietary sources combined. Internet resources, governmental entities, not-for-profit organizations, proprietary databases, and contacts at related entities are all utilized in order to gather the essential data for this step of the process.
When considering locations in the United States, project specifications can dictate that one particular region is evaluated more closely than another. States are eliminated based on a set of key criteria, including utility considerations, property tax rates, the timeline for securing permits in order to meet project construction deadlines, real estate acquisition costs, and level of local and/or state economic development incentives to support the project.
Somewhat unique to data centers, however, is the natural disaster factor. It is imperative for site selectors to understand the significant role geographic location plays for data centers. Natural disaster analysis must be conducted to ensure the risk factor is historically low for events such as earthquakes, floods, hurricanes, and tornadoes. This analysis will help the company that is considering where to locate a data center narrow down the initial list of potential sites in an efficient manner.
Once thorough information is collected and analyzed, an RFI or Request for Information is often sent to communities and states that are under consideration. Economic development organizations are asked to provide specific and relevant information only in their responses, and a quantitative analysis is done for each community under consideration. Utility costs, tax liability, and real estate and labor costs are all pieces of the thorough analysis that must be done in order to examine the projected operating costs - with and without factoring in the impact of any economic incentives.
The quantitative analysis compares each site under consideration with regard to energy costs, water rates, permitting fees, construction expenses, tax rates, labor quality and costs, and any other factors that are relevant and pertinent to the particular project's objectives. The quality-of-life factor is also important to ensure that the company can attract and retain the work force it needs to be successful. In addition, population numbers, educational levels, airport quality, climate, cost of living, crime rates, health care, commute times, and more are examined.
Once the number of locations has been narrowed down, personal visits are made to each location in order to meet with key local and state governmental authorities and economic development representatives. Initial negotiations begin regarding real estate acquisition, utility contracts, permitting timeline and costs, and economic development incentives issues are addressed at this stage. This allows the project team to narrow down the field to two or three potential location options.
Another piece of the puzzle comes into play, i.e., a qualitative analysis. In particular, the political stability of the selected finalist locations and any changes in key elected positions that could impact public policy are examined. The confidence in the community and state to deliver on promises/commitments is taken under consideration before a site selection recommendation is made. Trust, experience, and intuition - that "gut feeling" - are considered regarding the future direction of the community and state.
The site selection decision is based upon a detailed evaluation of the key project criteria and the assumptions for the project. The finalist locations are identified and summary information is gathered. In addition, the final package of economic development incentives for the project will have been negotiated before the site selection team makes its decision.
Economic incentives continue to play an important role in site selection decisions in today's world: data centers are no exception. Of course, a site selection decision should never be made based upon incentives. However, incentives are key differentiators at the final stage of the process. Offsetting project and operating costs for a data center facility through the effective use of incentives can be significant to the company making a facility location decision. Among the more common incentives that might be negotiated are tax credits, abatements, and exemptions; work force development grants and services; free or reduced-cost real estate; project financing (grants, loans or bonds); infrastructure grants; waiver of fees (impact, permit, and application); and utility costs reductions.
A Long-Term Decision
After all key issues related to the project are reviewed by key company decision-makers, they will be able to make an informed decision. Only after all project agreements are finalized with the appropriate governmental entities and the real estate is taken under control, should a public announcement of project plans be made.
It is important to never underestimate the level of research and analysis that is required to ensure a successful outcome. Staying focused on the desired outcome, analyzing the critical project criteria, using the key project assumptions, and communicating regularly with all parties involved in the process will help eliminate "surprises" down the road. Diligent research and analysis can result in operational savings and ensure the right long-term decision is made by the company. The end result is a highly efficient data center facility in a low-risk, cost-effective location.