Management of data storage and access are a part of every business, with a requirement for data centers and IT facilities common across nearly all business types around the world. Data Centers provide centralized IT systems, requiring power, cooling and operational requirements above and beyond typical design parameters. At 2009 figures, their demand for power accounted for approximately 0.5% of total global energy use McKinsey, 2009) making them one of the greatest sole consumers of energy and growing. This large density of power and cooling drives the need for continuous improvements and assessment of best practices in the industry.

One element for the design of data centers is the strategy used to control the conditions at the server racks and within the room. When CRAC units are used in a traditional arrangement, these can be operated using return air or supply air control systems, and the monitoring points can be located around the room or at the racks themselves. The selection of either control strategy leads to marked differences in energy used at the CRAC units and central chiller or cooling systems.

This paper provides a summary of the various methods for controlling data center conditions and will compare the energy performance per system using the Power Utilization Effectiveness (PUE) metric. It will include an assessment of any trends that are seen around the globe and determine if there are marked differences in operational acceptance and installer experience with each type.