Can energy efficiency programs keep up with data center growth?
This author no longer works for DNV GL.
Energy-intensive and growing rapidly, data centers offer unique challenges and opportunities for utility-sponsored energy efficiency programs. These facilities present significant potential for energy savings, but they don’t fit the usual prescriptive incentive approach. This article highlights a few challenges specific to new construction programs and recommends modifications to capture data center energy savings.
Data centers consume 10 to 50 times the energy per floor space of a typical commercial office building and account for approximately 2 percent of total electricity use in the U.S. The upward trend in energy use is increasing: data center loads are forecast to grow 10 to 15 percent annually due to increasing size and denser computing platforms.
DNV GL energy engineers and Data Center-Certified Energy Practitioners have reviewed numerous data center new construction and building upgrade projects and have performed site assessments for both small and large data centers. Our nationwide teams have processed more than 96 incentive applications for data center projects, surpassing $3.7 million in incentives paid and 49 GWh in annualized energy savings. Some of these projects have been dubbed the “greenest and coolest” by industry publications. Through this work, we have identified unique data center challenges and opportunities.What’s driving the need for more data center space? Global demand for personal electronics, exponential growth in mobile applications, an explosion in social media, storage needs for all that big data and the trend toward cloud computing, to name a few. As a result, data centers are becoming a larger and larger portion of many utility program portfolios.
Challenge 1: Data Center Baselines
Data centers are high-tech facilities with large process loads and specialized building systems. As a result, they should be held to a data center-specific baseline energy code. The American Society of Heating, Refrigeration and Air Conditioning Engineers (ASHRAE) has recognized the shortcomings in trying to apply commercial building energy codes such as ASHRAE 90.1 to data center facilities. ASHRAE 90.1 mainly addresses energy end-uses in buildings related to comfort conditioning for humans. It is inadequate for dealing with process-related loads like servers and mass storage devices found in data centers.
In response, ASHRAE is in the process of developing ASHRAE 90.4P Energy Standard for Data Centers and Telecommunications Buildings. Until ASHRAE 90.4P is released, programs should seek out alternatives to establish baseline equipment such as the Energy Efficiency Baselines for Data Centers report commissioned by a number of California utilities or by adopting a Power Utilization Efficiency (PUE)-based baseline.
Challenge 2: Analysis Methods
Regardless of which baseline is used, utility energy efficiency programs need to estimate the proposed building design’s energy usage. Many analysts use the same energy modeling software programs for data centers as they use standard commercial buildings; and these modeling software programs often can accurately estimate data center energy use. However, due to limitations in software, use of default software values, and a lack of understanding of data center building systems, these modeling software programs can also produce unrealistic energy estimates in large part because they are better suited for analyzing buildings predominately governed by comfort conditioning. Facilities with high process loads are problematic.
Challenge 3: IT Load Growth Rates
It may take a data center a few months or a few years to reach full design IT loads. Given these slow IT load growth rates, program implementers face the choice of paying an incentive based on the low IT loads at time of building completion or delaying payment until the IT load is higher. Neither option is ideal for the energy efficiency program or the data center. Alternatively, energy efficiency programs may consider phased, partial payments based on installed IT loads. Partial payments minimize program risk and increase program savings while providing the customer with the full incentive. Partial payments may not be possible for all programs and they do pose some added complexity, however, they can be a good option to increase program effectiveness and data center participation.
Data centers present a number of unique challenges to energy efficiency program implementation teams, but the payoff in potential energy savings is significant and the industry is forecast to continue its rapid growth. With increased awareness and by employing a few of our recommended modifications to existing approaches, utility programs can easily overcome these challenges.
Tarek Salameh is an Energy Engineer with DNV GL’s Technical Services team. Tarek provides engineering support to several energy efficiency implementation programs focusing on new construction. Tarek is a registered Professional Engineer in the State of Wisconsin and Certified Energy Manager.