Predicting data center energy costs and consumption is no easy task. But, with the right tools and expertise, it can be done.

Future energy costs are impacted by multiple variables, including overall energy consumption, the reliability and condition of your cooling and power systems, local utility rates and even the climate in which your datacenter is located, among others. These elements can be measured individually, but that doesn’t tell you the whole story. For an accurate picture of how much energy a data center consumes and what it will cost, you need to analyze all these variables together.

HP Critical Facilities Services (HP CFS), offers an innovative service to help companies get a more accurate assessment of data center energy cost and consumption. Through a combination of services and a proprietary energy assessment application, HP CFS can project a data center’s energy use by analyzing a broad range of variables simultaneously, and in relation to one another. For example, after entering all the relevant data variables into the energy assessment application, you can create “what-if” scenarios and run calculations to determine energy consumption and cost savings, as well as return on investment for power and cooling system upgrades in your facility.

This functionality helps you better understand an existing data center’s energy consumption and can provide a more clear picture of how proposed facility changes will play out. If you’re planning a new data center, it can even project energy costs and consumption for different geographic locations using climate and local utility rate data.

Most data centers maintain air temperatures much cooler than is necessary for IT equipment. See related article, “A more fuel-efficient data center”. In our example, the energy assessment application determined how much energy, and thus money, could be saved by allowing a higher temperature range in the data center.

In Graphic 1 below, you can see the difference in energy use over the course of one year between the data center’s current settings versus higher temperature settings. In the line graph, the solid line represents the data center’s current total energy use (base case), while the dotted line shows the potential energy use of increasing the temperature set points.

The bar graph in Graphic 1 compares the power usage effectiveness (PUE) of the two scenarios by each month of the year. PUE represents the ratio of total data center power use (including factors such as cooling) compared to the power use of your IT equipment alone. For example, a PUE of 2 means that your entire data center consumes two times the power of your IT equipment alone. So, the lower your PUE, the more energy efficient your data center.

Our hypothetical data center could save nearly three percent in energy costs just by setting the thermostat higher. In this case, that three percent could equal savings of more than USD$50,000 per year.

Adding a few key upgrades can produce even more significant results. Let's say our hypothetical company wants to make data center improvements, but isn’t sure where to focus. Based on the savings estimated by the energy assessment application, HP CFS recommended that the client replace its aging uninterruptible power supply (UPS) and corresponding electrical distribution.

Graphic 2 shows how a new UPS and electrical system along with a raised air temperature would affect power use compared with doing nothing and with a raised air temperature alone.

Climate is a major variable in a data center’s overall energy use because it affects how your mechanical systems perform (i.e., your cooling system).

Facilities in very warm climates require more energy to cool the data center, whereas those located in very dry climates require more energy to maintain humidity levels necessary to prevent electrostatic discharge. Key to HP CFS’ analysis is the assessment application's built-in climate and energy utility rate data for calculating energy costs for different locations around the world. This would allow our fictitious company to predict its potential energy use and costs in different locations and climates. It would be able to weigh those costs against other expenses and factors, such as real estate and proximity to metro centers, which are determined through other means.

Graphic 3 compares the climate trends of two potential locations with those of the current data center over the course of one year. The dotted line, which represents a location in London, shows the most moderate temperatures throughout the year, whereas the location in Hong Kong, represented by the dash and dot line, shows the highest overall temperatures year round. This assessment provides businesses with quantifiable data to evaluate the various locations and choose the one that’s most cost effective.

Of course, our hypothetical company could have done much more. But these three examples demonstrate how a data center energy assessment can not only save money, but help you plan for the future.

Whether you decide to make only minor adjustments to an existing facility, invest in new IT infrastructure or want to break ground and build a new data center, an energy assessment can play a vital role in your data center's (and your IT's) success. Learn more about how HP CFS can help.

	Your feedback is important to us. Was this article useful/informative?
	Not at all(1)		Neutral(3)		Definitely(5)