How to optimize data center energy efficiency with sensor selection 

Choosing the right HVAC sensors for data centers can have a significant impact on energy efficiency and power usage effectiveness (PUE). While sensor investment costs are relatively small, the impact on energy use can be disproportionately large. The right sensor solution may depend on the monitoring needs, and in the following article Anu Kätkä from Vaisala will discuss the issues that need to be considered when choosing sensors.

 

Small investment, big impact 

The number of data centers is growing, with over seven million worldwide in 2022 and a 10% annual industry growth rate. Data centers are incredibly energy intensive, accounting for more than 1% of global energy consumption – and, with increasing computing power needed, this figure is expected to increase. Therefore, the need to lower data center power consumption and improve sustainability is increasingly urgent. HVAC sensors have a significant role to play in addressing this challenge. 
 
Getting the balance right 

PUE is among the important figures to monitor when looking for improvements in energy efficiency. It is defined as the ratio of total energy used in a data center to the energy used solely by IT. Ideally PUE would be close to 1.0, which would mean that all energy is spent on IT and almost nothing on anything else. To achieve this it is necessary to minimize the consumption of supporting infrastructure, like cooling or power distribution. Historically, the typical PUE in a conventional data center has been around 2.0, whereas big hyperscale data centers have achieved a PUE below 1.2 or even less. In 2020 the global average PUE was 1.6, which would mean that on average 40% of energy was used for non-IT consumption.

 

As PUE is a ratio, it is a good idea to look at total consumption as well. It is also important to consider how the electricity is produced and whether renewable sources can be used, how the excess heat could be utilized, and how to reduce other environmental impacts like minimizing the amount of water used for cooling. In any case, one of the key drivers is minimizing energy consumption. The numbers are big: 1% of global energy consumed by data centers is a significant share, and up to 40% of that energy is used for cooling and air conditioning. Fortunately, that figure can be improved with accurate measurements. 
 
It all starts with a sensor 

Sensors and transmitters have an essential role in building automation and building management systems (BMS). Sensors are crucially important, because everything the system does is based on measurements. The BMS monitors prevailing indoor and outdoor conditions through parameters such as temperature and humidity, and performs controls. Precise indoor conditions can only be accurately maintained through accurate measurements, that can be trusted to remain accurate throughout the life of the building. 

 

Additional sensors may be necessary when a BMS is already in place and if there is a requirement to verify, monitor, and optimize the building systems – in this case a separate environmental monitoring system will be beneficial - to monitor, map, and show trends. It may also be necessary to report to third parties, for example to show that service level requirements have been fulfilled.  

 

High-quality sensors allow the optimization  of indoor conditions and the energy efficiency of a site, and thereby directly influence the PUE. This provides more control of the conditions in the data center, helping to improve indoor air quality, and the data center’s overall sustainability. 

 

Choosing a sensor solution 

There are several important criteria to consider when choosing a sensor solution, including reliability, stability, accuracy, and the needs of different spaces in the data center. Stability is the most important criterion; it means the sensor is accurate and reliable even in demanding environments, both now and in the long term. When buying sensors, the data sheets of different suppliers may state similar accuracies. It is easy to compare these numbers, but harder to understand and verify how accurate a sensor will be after one year, or a few years. All sensors drift, but some drift a lot or quickly, while others have good stability with minimal drifting, maintaining measurement accuracy for years. Therefore, the stability of sensors should be the most important criterion because it influences performance over a long time and is reflected in calibration and maintenance requirements.  

 

Despite good stability, high-quality instruments are also easy to maintain, calibrate, adjust, and fine-tune. Having convenient tools available for verifying and maintaining the sensors and transmitters provides reliability and peace of mind throughout the life cycle. High-quality, reliable, trustworthy sensors are assets that help optimize both the indoor environment and energy use. 

 

A sophisticated transmitter is also capable of providing calculated humidity parameters straight from the transmitter, for reliability and ease of system commissioning. Other factors to consider include modularity, exchangeable probes, and convenience – installation and commissioning should be flexible, while calibration and maintenance should be as easy as possible. A multi-use transmitter like the Vaisala HMT120 ticks all these boxes, allowing users to switch probes for calibration for example. The measurement probe is either attached to the transmitter itself, or attached by a cable, so the transmitter can be somewhere easy to read while the measurement is close to the process it needs to measure.

 

Optimized energy efficiency and lifetime value 

HVAC measurements have a critical role in control and monitoring, and measurement quality matters when it comes to PUE. Reliability and stability mean accuracy in the long term, which should be taken into consideration during instrument selection. In a nutshell, investing in high-quality measurement instruments just makes sense. It is a minor investment in the context of a data center project, but the value delivered by high-quality measurements is so much bigger. When users can trust the measurements, now and in the future, they are able to reach the best possible PUE and boost their sustainability performance. 

 

To find out more about optimizing data center performance with sensor selection, watch our webinar or get in touch.