LBNL Report Number
Eleven cooling systems for rack mounted computer equipment were evaluated and compared for their cooling effectiveness and their energy use. The cooling systems were of various designs but all were "close coupled" or "enclosed", and they all ultimately transferred heat to the building chilled water system. The types of cooling systems tested were: rack cooler with air to water heat exchanger, row type rack cooler with air to refrigerant or air to water heat exchanger, rack rear door passive cooler with air to refrigerant or air to water heat exchanger, a prototype direct touch cooling system using refrigerant, and a design consisting of a shipping container type enclosure cooled with chilled water.
A number of energy-efficiency metrics and test parameters were introduced and used for comparison. The energy-efficiency metrics follow industry-standard methods with some modifications. A set of three metrics calculates and compares energy efficiency using an algorithm of cooling provided divided by the electrical power needed. Another metric includes the power required for the rack-mounted electronic equipment, thus providing a metric for total power required.
All of the devices were effective in cooling the information technology (IT) equipment. The overall energy use efficiency varied considerably for a given cooling device configuration.