The Performance Indicator Assessing and Visualizing Data Center Cooling Performance

22 June, 2016 | White Paper

Editors: Mark Seymour, Future Facilities Maira Bana, Future Facilities David Wang, Teradata Danny Cummins, Siemens Veerendra Para, IBM Contributors: Magnus Herrlin, LBNL James Betts, Stulz Jay Vincent, Intel Henry Wong, Intel Chang Tsann, Dell Robery Langstrom, Interxion

The Performance Indicator introduced in this white paper presents three key cooling performance metrics (including one that utilizes Power Usage Effectiveness (PUE)) to build a balanced understanding of data center cooling performance.

The Green Grid developed the cooling performance metrics to ensure that, in the bid to save energy, the assessed facility maintains its ability to house and protect equipment during normal and resilient operation as it evolves throughout its life. This white paper presents definitions for the thermal conformance and thermal resilience of the IT as the cooling performance metrics. Using these metrics allows data center owners/operators to observe the compromises that may exist between maintaining acceptable thermal performance and striving for energy efficiency. It includes methods of calculation along with a detailed case study.

The white paper also discusses the fact that the Performance Indicator is not limited to these three metrics (or indeed to cooling performance alone) in the long term; rather, additional metrics can be accommodated as they are defined.

In order to obtain a complete and accurate assessment of a facility, it is necessary to calculate and display the metrics for the future states of the facility as well as its current state. The Green Grid recommends modeling, simulation, and power and temperature monitoring – all concepts this white paper addresses in detail.

Data centers are complex systems, and the various performance metrics are generally intertwined. The Performance Indicator allows the owner/operator to configure changes without excessive focus on a single metric, such as PUE, which could potentially result in stranded capacity or unintended risk.