Insight and analysis on the data center space from industry thought leaders.
The Impacts of Cooling and Energy Efficiency on Today’s Data Center Design
Before and after computational fluid dynamics (CFD) models can assist customers in predicting their energy savings and building an ROI model that improves the value of the containment investment.
October 11, 2016
Phillip Koblence is COO of NYI.
It’s 2016, yet IT experts are still challenged with how to effectively and efficiently cool their data center. The cooling process accounts for 40 percent of all power consumed by data centers, so this question is top of mind for operators. Ensuring optimal cooling in a data center not only lowers operational expenditure, but reduces the strain on equipment cooling mechanisms, extending the lifespan of the hardware; and freeing up power for IT equipment, increasing equipment uptime. The decision to invest in cooling infrastructure is easy, however choosing the method with which you regulate temperature within the data center can be more challenging.
Cooling and efficiency strategies are constantly evolving, with companies like Microsoft going so far as to drop a self-contained data center into the ocean. However, you do not need to plunge your equipment into the sea or move to the Arctic to keep yours cool. Hot-aisle containment (HAC) and cold-aisle containment (CAC) are the primary method in which leading businesses are reducing the use of energy and optimizing equipment performance within the data center. This proven and highly effective methodology of cooling has emerged as a new best practice within the industry.
Cooling and Energy Efficiency
Legacy data centers pose a challenge to containment due to the full separation of the supply and return airflow, which has evolved over time with the design of the room and equipment footprint. In these situations custom containment solutions are often required to address the issues plaguing the existing footprint, such as computer rows without opposing cold or hot aisles and isolated equipment that is not configured in a hot/cold aisle format.
HAC and CAC configurations within a data center ensure optimal efficiency by separating the cold supply airflow from the hot equipment exhaust air. By minimizing the mixing of hot and cold air, a uniform and predictable supply temperature is created and distributed to the intake of IT equipment. This results in a warmer, drier return air to the AC coil. By increasing air efficiencies, containment systems increase uptimes, extend hardware life and result in valuable cost and energy savings.
Furthermore, HAC and CAC containment systems are known for their ease of installation and limit the impact on existing infrastructure within a facility, which is important given floor space in a data center can be expensive. That said, it is important to understand the source of the supply airflow and return airflow when designing a HAC or CAC system. The room and equipment layout should dictate the design of the containment, dictating whether the aisle will be cold or hot. For instance, rooms with drop ceiling voids can be used to return air directly back to the CRAC without mixing with the room’s ambient airflow.
Containing Energy and Cost
Containment effectively reduces the power consumption of a data center and can result in significant cost benefits. The benefits can always be measured at the cooling intake of the servers. Before and after measurements show that the intake temperate can drop 10°. Additionally, return air temperatures increase 10°, which in turn increases the CRAC cooling capacity by over 50 percent.
The cost savings can be calculated directly from temperature within the data center. For every 1°F in temperature rise, the gain is 2 percent in cost savings for cooling. Additionally, containment allows data center providers to use the top 15 percent of the rack that is often left unpopulated due to inconsistent air supply temperatures, which results in an immediate savings that can run into the millions.
Before and after computational fluid dynamics (CFD) models can assist customers in predicting their energy savings and building an ROI model that improves the value of the containment investment.
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.
About the Author
You May Also Like