Skip to main content

Advertisement

SpringerLink
Log in

How do data centers make energy efficiency investment decisions? Qualitative evidence from focus groups and interviews

  • Original Article
  • Published:
Energy Efficiency Aims and scope Submit manuscript

Abstract

The data center industry is one of the fastest growing energy users in the USA. While the industry has improved its energy efficiency over the past decade, engineering analyses suggest that ample opportunities remain to reduce energy use that would save firms money. This study explores whether and why data centers might limit investment in energy efficiency. Given the scarcity of empirical data in this context, we conducted focus groups and interviews with data center managers to elicit information about factors affecting their investments and used content analysis to qualitatively evaluate the results. Split incentives between departments within companies and between colocation data centers and their tenants, imperfect information about the performance of new technologies, and tradeoffs with data center reliability were the most pervasive factors discussed by participants. While we find some evidence that market failure explanations such as split incentives and imperfect information had a limited role in slowing adoption for participants, rival explanations such as the cost of acquiring context-specific information, and opportunity costs associated with alternate uses of funds or highly valued attributes played a larger role in slowing investment in energy efficiency.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Explore related subjects

Discover the latest articles, news and stories from top researchers in related subjects.

Notes

  1. For a discussion of trends in data center energy efficiency in Europe, see Avgerinou et al. (2017).

  2. A 2013 California regulation mandates the use of specific energy-efficient cooling, airflow, and fan technologies in data centers (California Building Standards Commission 2013). Energy consumption for the rest of the industry remains unregulated in the USA, despite a few federal initiatives to promote best practices.

  3. The presence of unpriced negative externalities, such as greenhouse gases and other air pollutants, also leads to market failures, but externalities are not expected to be factored into firms’ private profit maximization decisions so we do not consider them here.

  4. A profit-maximizing firm invests until the additional benefit from the investment just equals the additional cost.

  5. The hypotheses we discuss are consistent with Sorrell et al. (2011) and Thollander et al. (2010).

  6. Conversely, it is possible that there are synergies between energy efficiency and other attributes, meaning that improvements in energy efficiency could actually lead to greater reliability or product quality in some cases. See “Tradeoffs between energy efficiency and other factors” section for a discussion in the context of data centers.

  7. A semi-structured approach uses a moderator’s guide, which lists the questions and topics to be discussed across interviews and focus groups. This approach allows the interviewer or moderator some discretion regarding the order in which the questions are asked and provides prompts to indicate when a topic is worthy of follow-up to facilitate a deeper understanding of a given topic as it relates to the hypotheses of interest.

  8. The questionnaire and moderator’s guide are available in an online appendix.

  9. Given the technical nature of the discussions and the fact that participants did not have access to questions ahead of time, we interpreted filler words—“um” or “err” or pauses—as indicative of a participant trying to gather his or her thoughts. These types of words were eliminated in the speech-to-text conversation process; we did not omit other informal speech mannerisms that might conceivably convey meaning.

  10. This form of coding is sometimes referred to as hypothesis coding and is a common approach to content analysis when attempting to assess researcher-generated hypotheses (Miles et al. 2014).

  11. We also used sub-codes. For instance, a discussion that was coded as “tradeoffs with other attributes” and “widely viewed as a slowing adoption” would also be tagged with “reliability” if it discussed how a technology affects performance, redundancy, uptime, or reliability.

  12. Focus groups (FG) and interviews (INT) are labeled numerically throughout the document.

  13. Four focus groups were held at Data Center Dynamics conferences in Dallas and New York, and two focus groups were held at an AFCOM conference in Boston.

  14. Excluding this sector could potentially bias our sample. However, the proportion of US servers in federal government data centers fell from roughly 5–10% in 2007 to 1–2% in 2012–2014 (Shehabi et al. 2016). Our sample does include private companies that provided computing services to federal government users.

  15. EPA and DOE’s Energy STAR buildings program recognizes top performers in energy efficiency through certification of individual facilities as well as portfolios of buildings or plants.

  16. The degree to which a data center operated as a colocation facility varied from less than 10 to over 90% of racks. More than half of the sample also leased space from others, though 25% leased a relatively small portion (i.e., 20% or less). In addition, most data centers occasionally outsourced facility management.

  17. Barriers that prevent mixing of incoming cold air and hot exhaust air (cold- or hot-aisle containment), tiles that guide cool air to servers (directional floor tiles, blanking panels), air conditioning unit devices that allow air flow to vary as cooling demand fluctuates (variable fan or speed drives), and higher temperature set points are frequently described as paying back in less than 2 years. Company statements and research indicate that investments in cooling method often have higher upfront costs, particularly retrofits, but can still yield large energy savings in some circumstances. See Van Geet (2017), Pacific West Air Conditioning (2014), Green Grid (2011), Wathaifi (2009), and https://www.energystar.gov/products/low_carbon_it_campaign/12_ways_save_energy_data_center/.

  18. This finding is consistent with Cooremans (2012) regarding the importance of investment characteristics in decision-making.

  19. A PUE of one indicates that a data center uses no additional power for lighting, cooling, power distribution, or facility operation beyond what is drawn by the IT equipment; a PUE of two indicates that for every unit of power consumed by IT equipment, another unit is used for facility operation. Participants reported a median PUE of 1.7 for localized facilities, 1.55 for enterprise facilities, and 1.4 for mega/utility-scale data centers.

  20. Howard and Holmes (2012) also found evidence of split incentives between the facilities and IT departments for the data centers they interviewed. They found that having one person responsible for both facilities and IT equipment decisions was more common among companies whose core business was data center services.

  21. While we do not know exactly how many participants rely on internal vs. external financing, discussion suggested they use a mix. Cooremans (2012) and Klemick et al. (2017) found that internal financing is prevalent in other sectors; as here, energy efficiency often competes with non-energy investments for internal capital.

  22. For an overview of incentive and rebate programs offered by utilities, see Howard and Holmes (2012).

  23. Mills et al. (2008) also pointed out that energy efficiency improvements in high-tech industries such as data centers often compete internally with other capital investments that pay off very quickly.

  24. These statements are consistent with Mills et al. (2008): risks of power failure are often overestimated, resulting in “overdesign” of backup systems and “strict emphasis on proven reliable technologies.” Howard and Holmes (2012) noted that a major challenge to improving data center energy efficiency is an “extreme focus on reliability.”

References

  • Andrews, C. J., & Krogmann, U. (2009). Explaining the adoption of energy-efficient technologies in U.S. commercial buildings. Energy and Buildings, 41, 287–294.

    Article  Google Scholar 

  • Avgerinou, M., Bertoldi, P., & Castellazzi, L. (2017). Trends in data centre energy consumption under the European code of conduct for data center energy efficiency. Energies, 10, 1470–1487.

    Article  Google Scholar 

  • Belady, C., Rawson, A., Pfleuger, J., & Cader, T. (2008). Green Grid data center power efficiency metrics: PUE and DCIE. White Paper No. 6. Green Grid. https://www.thegreengrid.org/en/resources/library-andtools. Accessed 22 Mar 2019.

  • Bennet, D., & Delforge, P. (2012). Small server rooms, big energy savings. National Resource Defense Council (NRDC) Issue Paper. February. NRDC. https://www.nrdc.org/sites/default/files/Saving-Energy-Server-Rooms-IssuePaper.pdf. Accessed 22 Mar 2019.

  • Brown, R., Masanet, E., Nordman, B., Tschudi, W., Shehabi, A., Stanley, J., et al. (2007). Report to congress on server and data center energy efficiency: public law 109-431. Lawrence Berkeley National Laboratory. LBNL-363E. http://eta-publications.lbl.gov/sites/default/files/pdf_3.pdf. Accessed 2 Mar 2019.

  • Brunke, J.-C., Johansson, M., & Thollander, P. (2014). Empirical investigation of barriers and drivers to the adoption of energy conservation measures, energy management practices and energy services in the Swedish iron and steel industry. Journal of Cleaner Production, 84, 509–525.

    Article  Google Scholar 

  • Bullock, M. (2009). Data center definitions and solutions. CIO Digital Magazine. August 14. https://www.cio.com/article/2425545/data-center/data-center-definition-and-solutions.html. Accessed 22 Mar 2019.

  • California Building Standards Commission. 2013. Callifornia Energy Code 2013. California Code of Regulations Title 24, Part 6. Available at: https://archive.org/stream/gov.ca.bsc.2013.6/#page/n0/mode/2up

  • Cisco (2016). Global cloud index: forecast and methodology, 2015–2020. Cisco Public White Paper. https://www.cisco.com/c/dam/m/en_us/service-provider/ciscoknowledgenetwork/files/622_11_15-16-Cisco_GCI_CKN_2015-2020_AMER_EMEAR_NOV2016.pdf. Accessed 22 Mar 2019.

  • Cooremans, C. (2011). Make it strategic! Financial investment logic is not enough. Energy Efficiency, 4, 473–492.

    Article  Google Scholar 

  • Cooremans, C. (2012). Investment in energy efficiency: do the characteristics of investments matter? Energy Efficiency, 5, 497–518.

    Article  Google Scholar 

  • Dixon-O’Mara, C., & Ryan, L. (2018). Energy efficiency in the food retail sector: barriers, drivers, and acceptable policies. Energy Efficiency, 11(2), 445-464.

  • Frechtling, J., and L. Sharp, eds. 1997. User-Friendly Handbook for Mixed Method Evaluations. Produced for the National Science Foundation

  • Gerarden, T., Newell, R., & Stavins, R. (2017). Assessing the energy-efficiency gap. Journal of Economic Literature, 55(4), 1486–1525.

    Article  Google Scholar 

  • Gillingham, K., Newell, R., & Palmer, K. (2009). Energy efficiency economics and policy. Annual Review of Resource Economics, 1, 597–620.

    Article  Google Scholar 

  • Green Grid (2011). Case study: the ROI of cooling system energy efficiency upgrades. White paper #39. https://www.energystar.gov/sites/default/files/asset/document/CaseStudy_TheROIofCoolingSystemEnergyEfficiencyUpgrades.pdf. Accessed 22 Mar 2019.

  • Gul, M., & Menzies, G. (2012). Designing domestic buildings for future summers: attitudes and opinions of building professionals. Energy Policy, 45, 752–761.

    Article  Google Scholar 

  • Howard, A., & Holmes, J. (2012). Addressing data center efficiency: lessons learned from process evaluations of utility energy efficiency programs. Energy Efficiency, 5, 137–148.

    Article  Google Scholar 

  • Huntington, H. 2011. “The Policy Implications of Energy-Efficiency Curves.” Energy Journal, Special Issue: 7–22.

  • Hsieh, H.-F., & Shannon, S. (2005). Three approaches to qualitative content analysis. Qualitative Health Research, 15(9), 1277–1288.

    Article  Google Scholar 

  • Jaccard, M. 2010. “Paradigms of Energy Efficiency's Cost and Their Policy Implications: Deja Vu All Over Again.” Modeling the Economics of Greenhouse Gas Mitigation: Summary of a Workshop. National Research Council. National Academies Press. Washington, D.C.

  • Jaffe, A., & Stavins, R. (1994). The energy-efficiency gap: What does it mean? Energy Policy, 22, 804–810.

    Article  Google Scholar 

  • JLL (2016). Data center outlook: strong demand, smart growth. North America. http://www.us.jll.com/united-states/en-us/Research/US-North-America-Data-Center-Outlook-2016-JLL.PDF. Accessed 22 Mar 2019.

  • Klemick, H., Kopits, E., Wolverton, A., & Sargent, K. (2015). Heavy-duty trucking and the energy efficiency paradox: evidence from focus groups and interviews. Transportation Research Part A: Policy and Practice, 77, 154–166.

    Google Scholar 

  • Klemick, H., Kopits, E., & Wolverton, A. (2017). Potential barriers to improving energy efficiency in commercial buildings: the case of supermarket refrigeration. Journal of Benefit Cost Analysis, 8(1), 1–31.

    Article  Google Scholar 

  • McKinsey & Company (2009). “Unlocking Energy Efficiency in the U.S. Economy.” Report. McKinsey. https://www.mckinsey.com/~/media/mckinsey/dotcom/client_service/epng/pdfs/unlocking%20energy%20efficiency/us_energy_efficiency_exc_summary.ashx. Accessed 22 Mar 2019.

  • Miles, M., Huberman, A., & Saldaña, J. (2014). Qualitative data analysis: a methods sourcebook (Third ed.). Thousand Oaks: Sage Publications.

    Google Scholar 

  • Mills, E., Shamshoian, G., Blazek, M., Naughton, P., Seese, R., Tschudi, W., et al. (2008). The business case for energy management in high-tech industries. Energy Efficiency, 1, 5–20.

    Article  Google Scholar 

  • National Resources Defense Council (NRDC) (2014). Scaling up energy efficiency across the data center industry: Evaluating key drivers and barriers. Issue Paper IP:14-08-a. August. https://www.nrdc.org/sites/default/files/data-center-efficiency-assessment-IP.pdf. Accessed 22 Mar 2019.

  • Ochieng, E., Jones, N., Price, A., Ruan, X., Egbu, C., & Zuofa, T. (2014). Integration of energy efficient technologies in UK supermarkets. Energy Policy, 67, 388–393.

    Article  Google Scholar 

  • Office of Management and Budget (OMB) (2003). “Circular A-4” September 17. Available at: https://www.whitehouse.gov/sites/whitehouse.gov/files/omb/circulars/A4/a-4.pdf. Accessed 22 Mar 2019.

  • Olson, M. (1965). The logic of collective action: public goods and the theory of groups. Cambridge: Harvard University Press.

    Google Scholar 

  • Pacific West Air Conditioning. 2014. “Water-side economizers.” November. Available at: http://www.pacwestac.com/energy-efficiency-water-side-economizer.html. Accessed 22 Mar 2019.

  • Palm, J., & Reindl, K. (2018). Understanding barriers to energy-efficiency renovations of multifamily dwellings. Energy Efficiency, 11, 53–65.

    Article  Google Scholar 

  • Pellegrini-Masini, G., & Leishman, C. (2011). The role of corporate reputation and employees’ values in the uptake of energy efficiency in office buildings. Energy Policy, 39, 5409–5419.

    Article  Google Scholar 

  • Radcliff, B., & Best, S. (Eds.). (2005). Polling America: an encyclopedia of public opinion research. Westport: Greenwood Press.

    Google Scholar 

  • Schleich, J., & Gruber, E. (2008). Beyond case studies: barriers to energy efficiency in commerce and the services sector. Energy Economics, 30, 449–464.

    Article  Google Scholar 

  • Shehabi, A., Smith, S., Sartor, D., Brown, R., Herrlin, M., Koomey, J., et al. (2016). United States data center energy usage report. Lawrence Berkeley National Laboratory. LBNL-1005775. http://etapublications.lbl.gov/sites/default/files/lbnl-1005775_v2.pdf Accessed 22 Mar 2019.

  • Sorrell, S., Mallett, A., & Nye, S. (2011). Barriers to industrial energy efficiency: a literature review. United Nations Industrial Development Organization. Working Paper 10/2011.

  • Sullivan, R., & Gouldson, A. (2013). Ten years of corporate action on climate change: what do we have to show for it? Energy Policy, 60, 733–740.

    Article  Google Scholar 

  • Thollander, P., & Ottosson, M. (2008). An energy efficient Swedish pulp and paper industry – exploring barriers to and driving forces for cost-effective energy efficiency investments. Energy Efficiency, 1, 21–34.

    Article  Google Scholar 

  • Thollander, P., Palm, J., & Rohdin, P. (2010). Categorizing barriers to energy efficiency – an interdisciplinary perspective. In J. Palm (Ed.), Energy Efficiency. INTECH (Open Access). https://www.intechopen.com/books/energy-efficiency/categorizing-barriers-to-energy-efficiency-aninterdisciplinary-perspective Accessed 22 Mar 2019.

  • Van Geet, O. (2017). A method for estimating potential energy and cost savings for cooling existing data centers. National Renewable Energy Laboratory (NREL). https://www.nrel.gov/docs/fy17osti/68218.pdf. Accessed 22 Mar 2019.

  • Wathaifi, M. H. (2009). Free cooling using water-side economizer: process cooling application at a plastic extrusion plant. Energy Engineering, 106(6), 45–78.

    Article  Google Scholar 

  • Yin, R. K. (2003). Case study research: design and methods (Third ed.). Thousand Oaks: Sage Publications.

    Google Scholar 

Download references

Acknowledgements

The authors thank three anonymous reviewers for their comments, as well as Barbara Bauer and David Cooley (Abt Associates); Linda Dethman and Jane Peters (Research Into Action); Beth Binns, Datacenter Dynamics, AFCOM, Keith Sargent, and Cynthia Morgan for help with focus groups and interviews and for useful input.

Funding

Focus groups and interviews were conducted with contractor support funded by the U.S. EPA.

Author information

Authors and Affiliations

  1. U.S. EPA, 1200 Pennsylvania Ave, NW, MC 1809T, Washington, DC, 20460, USA

    Heather Klemick, Elizabeth Kopits & Ann Wolverton

Authors
  1. Heather Klemick
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Elizabeth Kopits
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Ann Wolverton
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ann Wolverton.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Disclaimer

All opinions expressed in the paper are those of the authors and do not necessarily reflect the views or policies of the U.S. Environmental Protection Agency (EPA).

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

ESM 1

(DOCX 37 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Klemick, H., Kopits, E. & Wolverton, A. How do data centers make energy efficiency investment decisions? Qualitative evidence from focus groups and interviews. Energy Efficiency 12, 1359–1377 (2019). https://doi.org/10.1007/s12053-019-09782-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s12053-019-09782-2

Keywords

Search

Navigation