Impact of Hot and Cold Aisle Containment on Data Center Temperature and Efficiency

by John Niemann, Kevin Brown, and Victor Avelar

Executive Summary

Both hot-air and cold-air containment can improve the predictability and efficiency of traditional data center cooling systems. While both approaches minimize the mixing of hot and cold air, there are practical differences in implementation and operation that have significant consequences on work environment conditions, PUE, and economizer mode hours. The choice of hotaisle containment over cold-aisle containment can save 43% in annual cooling system energy cost, corresponding to a 15% reduction in annualized PUE. This paper examines both methodologies and highlights the reasons why hot-aisle containment emerges as the preferred best practice for new data centers.

Conclusion

Prevention of hot and cold air mixing is a key to all efficient data center cooling strategies. Both HACS and CACS offer improved power density and efficiency when compared with traditional cooling approaches. A hot-aisle containment system (HACS) is a more efficient approach than a cold-aisle containment system (CACS) because it allows higher hot aisle temperatures and increased chilled water temperatures which results in increased economizer mode hours and significant electrical cost savings. Cooling set points can be set higher while still maintaining a comfortable temperature in the uncontained area of the data center. The analysis in this paper shows that HACS can save 43% in the annual cooling system energy cost corresponding to 15% reduction in the annualized PUE compared to CACS while holding the uncontained data center area to 24?C/75?F. This paper concludes that all new data center designs should use HACS as the default containment strategy. In cases where containment is not initially required, the new data center design should incorporate provisions for future HACS deployment. For existing raised floor data centers with a perimeter cooling unit layout, it may be easier and less costly to implement CACS.