Conserving data center energy is now becoming a top priority for data center managers across the world. Thus, this has made the server facility managers to keep a track on facility’s total power usage against the power used by IT equipment. And to address the need for an industry-wide benchmark, The Green Grid developed the power usage effectiveness (PUE) calculation in 2007 as a principle way to measure the efficiency of data center infrastructure.
But before embarking on their PUE measurement strategy, data center managers must clarify three things in a clear cut way. 1.) There must be an agreement on exactly what devices constitute to IT load; 2.) What devices constitute physical infrastructure and what devices should be excluded from the measurement?
If these three factors remain unsolved, then those managing data centers can never get accuracy over PUE. However, this process is easier said than done, as the following issues have to encounter at first.
It is evident that all electronic devices draw power. But one must know the exact way to count that power drawing data in PUE calculations as-
- Various power consuming data center subsystems such as outdoor lighting or the Network Operation Center aren’t represented in data center power calculations.
- Some subsystems (Chiller plants and cooling towers) support mixed-use facility and are shared with other non-data center functions. So fractions of the power attributable to the data center cannot be directly measured.
- Some practical power measurements points include loads that are unrelated to data center, but cannot be separated from the measurement.
- Therefore, usually all these parameters are not considered while calculating the PUE of data centers, as these are often ignored by most data centers managers from standard methodology. Thus, the same data center can have a different energy efficiency rating when different parameters are taken into consideration.
So, what should the data center or facility manager do?
The above said issues can create trouble to data center operators who are attempting to calculate data center PUE.
Therefore, it is better to approach in the following way to determine data center PUE in an affective way. Also the below said methodology defines a standard approach for collecting data and drawing insight from data centers. It also helps data center managers understand how to use this approach to calculate PUE, with a focus on what to do with data that is either misleading or incomplete.
First establish a standard for differentiating physical infrastructure and IT loads
In going with this methodology, one can categorize subsystems as either 1.) IT loads or 2.) Physical infrastructure or 3.) Determine whether the subsystem should be excluded in the calculation. For instance, servers and storage constitute IT loads and UPS and HVAC constitute physical infrastructure. Subsystems are those like personnel spaces, switchgear, and the NOC, which all draw power, but cannot be categorized into physical infrastructure or IT loads. But if these systems are ignored while calculating power efficiency of a data center, then the whole PUE calculations can fall into doldrums and can be of no use. Since many customers, government bodies, utilities, and data center providers are looking for a standard benchmark for data center efficiency, clear guidelines for what is classified as an IT load or physical infrastructure are critical to determining a benchmark that can used across various data centers.
Include shares resources in PUE calculations-
Some devices which reside in data center facilities consume power. But are also shared for other uses such as a chiller plant or a UPS that also offers power backup to front office, a call center and as such. Thus, these devices cannot be termed directly as power resource consumers of a data center. But avoiding it in PUE calculations of a data center can cause major errors, especially if the device is a major energy user like a chiller plant.
Therefore, a better way to include these shared resources into the PUE calculations is to take the fraction of losses that are associated with data center, and then use those losses to determine the PUE.
If suppose a chiller plant can be termed as a shared resource in your facility, then
Measure/estimate the thermal load on the chiller using all the electrical losses of all the other data center loads, then estimate the chiller performance. This will help in determining a good estimate of how much of the chiller’s power the data center is using.
Measure the fractional split of the thermal load between the data center and the other loads. Using water temperature, pressure, pump setting, etc., measure the chiller input power, and then allocate the traction of the chiller power to the data center according to the fractional split.
Shut off the non-data center loads on the chiller, and then measure it to determine the power offset for the data center. These indirect estimates are best made during an expert data center energy audit, and once the technique is established it can frequently be used over time when efficiency trending is important
Better to guestimate the devices that are impractical to measure-
Factually speaking, we can measure the power consumption of all electrical devices. But in some cases, power consumption estimation for some devices becomes impractical. For instance, consider a power distribution unit in a partially loaded data center. Its losses can be in excess of 10% of IT load. These loss figures can significantly impact PUE, yet most data center operations omit PDU losses in PUE calculations because they can be difficult to determine when using the built-in PDU instrumentation. Thus, a tough nut to crack…..isn’t it?
However, Victor Avelar—a Senior Research Analyst for Schneider Electric’s Data Science Center, provides a calculative solution for PDU losses. Victor feels that the loss can be directly calculated from the IT load with precise accuracy, if the load is known in watts, amps or VA. Victor formulates that once the estimated PDU losses are subtracted from the UPS output metering to obtain the IT load; they can be counted as a part of the infrastructure load. This method improves the PUE calculation, as opposed to ignoring PDU.
Hope, the above said three methodologies help facility managers calculate accurately and effectively their data center PUE to ensure that their computing facilities meet not only internationally recognized energy efficiency regulations, but also larger business goals in competitive spirits.