Being in the loop

All about the Industry and Submer Tools

August 24, 2021

# 3 Metrics for Datacenter Efficiency: PUE, CUE and WUE

Three metrics, measuring three different factors. Which is the most important for the Datacenter industry?

PUE, CUE and WUE metrics

Like any industry, Datacenters have a set of metrics that help to set the benchmark for its success in a number of areas, including efficiency, sustainability and density. The metrics are defined as Power Usage Effectiveness (PUE), Carbon Usage Effectiveness (CUE) and Water Usage Effectiveness (WUE), As the business of Datacenters continues to experience exponential growth without signs of slowing down, there is growing debate about what metrics best measure which Datacenters are currently performing the best across the board. Let’s take a closer look at three of the main metrics below.

1. Power Usage Effectiveness (PUE)

Power Usage Effectiveness (PUE) has been widely employed in the research, development and evaluation of data centers as one of the most important metrics to measure systems performance and efficiency. It is defined as the total energy used by a data center against the energy employed only by its IT equipment:

PUE metric equation

Where Non-IT Equipment Energy is essentially referred to as operational and infrastructure costs. This metric is attractive due to its simplicity, its meaning in terms of energy consumption and its versatility to evaluate a whole system or different sections of a data center facility. Nevertheless, in a dynamic world with an ever-growing environmental concern, new metrics could be potential game-changers in the design trends of the next generations of data centers.

Let’s take a look at the possible alternatives.

2. Carbon Usage Effectiveness (CUE)

Carbon Usage Effectiveness (CUE) aims to measure its sustainability by means of pollutant emissions as a consequence. CUE (carbon usage effectiveness), is then defined as the relation between the CO2 emissions produced by the datacenter and the energy consumption of IT equipment:

CUE equation

Whereas Carbon Dioxide Emission Factor (CEF) specifies the CO2 factor of the electrical power. This value can be country-specific or even system-specific depending on the respective mix of electrical power sources (coal, nuclear, gas, wind, etc.). Therefore, in an environmentally ideal scenario where a datacenter is designed to work with 100% renewable electricity, CUE values theoretically are equal to ‘0’.

3. Water Usage Effectiveness (WUE)

Water Usage Effectiveness (WUE) is an indicator defined as the ratio between the use of water in a datacenter system (water loops, adiabatic towers, humidification, water-driven energy production, etc.) and the energy consumption of the IT components:

WUE metric equation

While this definition is site-based, the water needed for the production of the energy that powers the datacenter is a different aspect to take into account. Estimations on U.S. data center energy usage [Shehabi et al., 2016] determined that far more water is used for power generation (an average of 7.6 liters per kWh) than for IT cooling (1.8 liters per kWh). This gives an additional dimension to the need to keep PUE (that consequently affects a source-based WUE) as low as possible. Facebook, one of the leaders of the industry, has focused efforts to prioritize smart water use and reported an average WUE of 0.24 liters per kWh in their data centers [Facebook, 2020]. Nevertheless, this is still a relatively new metric that is rarely reported by the majority of companies in the sector.

The increased usage of water, even if favorable for power consumption stats, may lead to huge water bills alongside the environmental challenge of water becoming limited in the future due to climate change (40% gap between global water supply and demand by 2030 [McKinsey, 2009]). Many data centers are becoming aware of this issue and use WUE to evaluate their water consumption.

How will PUE, CUE and WUE metrics affect the future of the Datacenter industry?

Along with the business benefits of energy savings, the exponential use of data and the increase in public and legislative awareness in environmental issues have placed increased pressure on data center designers and companies to follow green policies. Then, WUE and CUE are expected to become essential key performance indicators for data centers, in addition to the extensively used PUE. This raises the need for the implementation of innovative IT cooling solutions against traditional alternatives.

How can Submer help improve a Datacenter’s PUE, CUE and WUE metrics

One of the most promising technologies in the pursuit of ever-more efficient data centers is Immersion cooling. Submer, one of the strongest pioneers developing these kinds of systems, has developed and deployed multiple solutions that incorporate its dielectric coolant (SmartCoolant) that not only have zero water consumption due to their management of coolant circuits, but also a certified PUE of 1.03. This reduction in PUE not only represents unprecedented energy savings but, in consequence, a significant reduction in source-based WUE, in a world in which energy generation is still far from being entirely sustainable.  

To find out more about how Submer can help you reduce key power and efficiency metrics to enable a more smart datacenter, take a look at our solutions here.