This site uses cookies. By continuing to browse the site you are agreeing to our use of cookies. Read our privacy policy (update in May 2018) >

DC PUE Optimization

Pain Points

Data center power usage effectiveness (DC PUE) is the energy efficiency of a data center. A PUE calculation method is total energy consumption of the data center/IT load power consumption. The DC PUE is usually between 1.4 and 1.5. In an example in which the annual average PUE of a data center is 1.43, the electricity fees account for 70% of the total operation cost, and the electricity fees of the cooling system accounts for 19% of the total electricity fees. Therefore, reducing the energy consumption of the cooling system is the goal of reducing the DC PUE. Although the current cooling control system has been automated, it can only perform automatic operations, for example, adding or reducing units, after users set control policies based on their experience. This method cannot obviously reduce the DC PUE, cannot analyze the characteristic parameters affecting the DC PUE and build a model to optimize global parameters, and cannot achieve a refined operating control policy for cooling towers, cooling devices, and water pumps.


Solution Benefits

By introducing artificial intelligence (AI) into the cooling control system of the data center, you can find the optimal combination from millions of parameter combinations, to reduce the energy consumption of the cooling system and reduce the operating expense (OPEX). The application result of a data center in city L in province H, China shows that the energy efficiency is improved by 15% based on the AI technology, and the electricity fees are reduced by RMB 3 million/year.