How Does Warm Weather Impact Your Data Centre? Challenges & Strategies

by | Aug 4, 2023 | Articles, Best Practice, Efficiency & Optimisation

In ever-unpredictable climates and warm weather spells, we are becoming even more conscious that data centres can be significantly affected by hot weather conditions. This is a growing concern as global temperatures continue to rise.

Here are some of the ways hot weather can impact data centres:

Cooling Challenges

IT Equipment hosted within data centres generates a considerable amount of heat due to the constant operation of servers and other hardware. Broadly speaking, all the power delivered to the IT equipment is converted to heat.

To maintain optimal operating temperatures and prevent hardware failure, data centres require efficient and reliable cooling systems to reject this heat.

In hot weather, cooling systems must work harder to dissipate heat, leading to increased energy consumption, potential cooling capacity issues, and higher equipment failure rates.

If your cooling infrastructure is old, poorly maintained, or close to capacity, your risk of failure and highly disruptive data centre outages increases significantly.

Cooling Plant Challenges

Many cooling plants in the UK data centre market have been sized and deployed based on 10-year UK weather data and are meant to last their end user up to 15 years, if well maintained.

What historical weather data doesn’t provide you with, is projected new highs in temperature that are now being anticipated – this is where the problems begin.

If your cooling plant is already past the end of life, it may be struggling to perform even on 30°C degree days so the risk of complete failure at higher temperatures is very real.

Systems that have been installed over the last few years were often designed to operate at a maximum temperature of 34°C to 36°C, so what’s going to happen when it exceeds this temperature?

Even if you have a cooling plant deployed in the last 5 years with consideration for higher maximum temperatures,

  • Has it been installed correctly?
  • Has it been properly commissioned?
  • Was it sized correctly in terms of capacity?
  • Has it been maintained correctly?
  • Is there sufficient air movement around the external heat rejection equipment to prevent temperature pooling?
  • Are refrigerant levels sufficient?
  • Do set points need to be adjusted for the summer?
  • Are radiating surfaces clean and not damaged or corroded?
  • Are fans all functioning correctly and obstruction free?
  • Are filters clean and when were they last replaced?

There is a lot to consider as all these factors and more will have a direct impact on the ability of your cooling system to support your critical load.

Image | future tech

Overheating and Hardware Failures

If data centre cooling systems are overwhelmed by high external temperatures during periods of unusually hot weather, internal temperatures will very rapidly rise to critical levels.

This can result in hardware failures, system crashes, and data loss. Even in mild cases, high temperatures will shorten the lifespan of components, leading to increased maintenance and replacement costs, as well as the potential for loss of equipment warranties.

In addition, there is the risk of customer SLA breaches with the result that financial penalties will be imposed along with loss of customer confidence and reputational damage.

In more extreme cases, critical failure of the entire facility will also be experienced. It is at this point that the wider business, and those not directly involved in data centre operations day-to-day, realise just how critical, reliable, and highly available data centre operations are to the business.  

Power Usage and Energy Costs

As data centres cope with increased cooling demands during hot weather, the power usage of inefficient cooling systems rises significantly. This will lead to higher energy costs for data centre operators and contributes to overall energy consumption and greenhouse gas emissions.

With the increasing focus on ESG reporting this is something that both data centre operators and their customers seek to avoid.

Reduced Efficiency and Performance

High temperatures can negatively impact the efficiency and performance of data centre equipment.

UPS VRLA battery life will decrease dramatically when temperatures are consistently higher than 25°C; compressors and expansion valves can fail; servers may need to throttle their processing speed to prevent overheating or shut themselves down entirely to avoid damage.

All these can lead to reduced function in your data centre, a reduction in services and service quality based on computing capabilities, and higher equipment failure rates.

Risk of Outages

In extreme cases, prolonged hot weather conditions can overload data centre cooling systems to the point of failure, leading to unplanned outages.

These outages can be costly for businesses that rely on data centre services and may result in downtime, customer dissatisfaction, loss of revenue, loss of reputation, SLA penalties, and in the worst cases a complete loss of business operations.

Many businesses simply do not know how dependent they are on continually available data centres and how critical the data centre is to their business operations until they have a failure.

Water Availability

Some data centres use water-based cooling methods, which rely on a stable and sufficient water supply. During periods of hot weather, water scarcity or restrictions on water usage may impact the cooling capabilities of these data centres.

To address these challenges and mitigate the impact of hot weather on data centres, operators are implementing a range of strategies:

Planned Preventative Maintenance

It is critical to consider the ongoing operation of your data centre rather than simply the way it is designed to operate – you don’t just deploy once and sit back.

Be it on-premises or in third-party space off-premises, a well-maintained facility will match expectations and continue to perform according to the design intentions. A comprehensive and well-considered maintenance plan is essential to achieving this.

You would not expect your car to continue to run perfectly according to the manufacturer’s specifications and performance guidelines, or even potentially start every time if you did not keep it well maintained. Data Centres are no different.

Improved Cooling Systems

Believe it or not, not all air conditioning is born equal. AC, HVAC, and CRAH are all terms used to describe cooling plants.

Data centres are adopting more advanced and energy-efficient cooling technologies, such as free cooling, immersed liquid cooling, on-chip cooling and simpler quick wins such as hot/cold aisle containment, adiabatic options such as indirect (and direct), evaporative cooling to better handle heat load, and to improve energy efficiency during hot weather.

If you want to know about any of these innovative methods and more, please contact Future-tech

Location and Design

When planning new data centres or expanding existing ones, it is very important to engage with the right people who are genuine experts in data centre engineering.

Technical decisions are frequently based on cost without fully understanding the impacts of so-called ‘Value Engineering’. Employing a cheaper solution may have unforeseen business impacts down the line.

Designs will also be affected by location. What might work well in one location may not be right for another, so it is important that operators consider the location, local climate (high and low ambient temperatures), altitude, and humidity, and ideally opt for locations with suitable ambient conditions to improve both reliability and energy performance.

The right location allows data centres to be designed to maximise the use of energy-efficient cooling methods, based on compressor-free cooling to take advantage of favourable external ambient air conditions to significantly reduce cooling power consumption, where the local environment is right to do so.

Image 1 | future tech

Thermal Monitoring and AI

Data centre operators can increasingly use sophisticated monitoring systems and AI algorithms to detect temperature fluctuations and make suggestions for real-time adjustments to cooling systems.

This ensures optimal cooling and energy usage performance, reduces the risk of ‘hot spots’, simplifies cooling capacity planning, eliminates stranded cooling capacity, and ultimately eliminates the risk of overheating.

As the frequency and intensity of heat waves and extreme weather events increase, data centres and the IT industry in general are taking steps to adapt and become more resilient to these climate challenges, all while at the same time seeking to improve energy efficiency.

However, many people are simply burying their heads in the sand during the cooler months, in the belief that it is going to be prohibitively expensive to replace the cooling plant.

This is not always the case.

Sometimes operational changes within the data centre and optimisation of the cooling within the IT spaces can help, but it’s not always about simply throwing more cooling at the problem as that may not always help.

If you are concerned or would like more information on cooling options, cooling optimisation or to better understand any potential risks you might be facing, please contact Future-tech and we will be happy to help.