The mercury is rising this week as hot air arrives from the continent, and with rising global temperatures, effectively cooling data centres becomes ever more vital. Technological advancements and reliance on innovations such as 5G and AI across industries means that downtime due to cooling related system failures must be avoided.
Uptime Institute’s 2021 annual outage analysis, lists cooling failures in the top 3 causes of significant outages in data centres, so it is clear that contractors and consultants need to ensure procedures and infrastructure are in place from the construction phase onwards to protect against cooling system failures.
In this article, we explore the impact climate change is having on cooling and the legislation that contractors and consultants in the sector must be aware of.
A data centre’s cooling infrastructure is critical and there are a variety of methods that are more suited to particular locations or uses, whether that be air or liquid cooling. Usually found in older data centres, mechanical HVAC systems for cooling can be relatively inefficient when it comes to energy consumption, leaving some providers and operators turning to alternative methods.
Given that the data centre industry consumes over 200TWh of electricity a year, cooling infrastructure is also required to be as efficient as possible to keep up with stringent environmental targets set by governments and tech companies. An efficient means of non-mechanical temperature control, free cooling uses fresh air blended with recirculating air form within the data centre to efficiently cool the server spaces.
Particularly in regions where climates are colder, this method harnesses the already chilled air and helps in the reduction of energy consumption. As the environmental impact of data centres continues to be scrutinised, for the UK and Ireland as well as colder climates such as the Nordics region, this plays a key role in delivering greener data centres.
However, facilities that use free cooling may find that, during summer months, their systems could be limited when coping with increasing temperatures or heatwaves. Free cooling can begin to reach its threshold as the ambient temperature reaches above 25oC at which point the server rooms risk becoming overheated. With the risk of downtime looming over data centre professionals, onus is falling on designers and consultants to help mitigate the situation from early on in the lifecycle.
Affecting older data centres with mechanical cooling systems, the 2020 F-Gas ban is a requirement of EU Regulation 517/2014 and means no refrigerants with a high global warming potential (GWP) can be used. Though many facilities may change over to a free cooled approach, older data centres with more equipment introduced when they were first commissioned, require refurbishment or replacement of aging infrastructure to be compliant.
Planned maintenance schedules for data centres across Europe, need to ensure that cooling systems are compliant with regulations after any work is carried out. Continuing supply chain issues may mean that this process could be prolonged while waiting for equipment to arrive, leaving further possibility for equipment failure.
For data centres, the huge cost and reputational damage that comes with periods of downtime means stringent planning for the mitigation of all potential pitfalls is vital. Specifying the right temperature control system can be a highly technical task, especially with variations in sizes and requirements of different data centres. With this in mind it is vital that a robust contingency plan with all of the intricate requirements is in place ready to implement when temperatures climb.
This type of planning can be implemented by operators and providers to mitigate all potential pitfalls, from power outages to potential supply chain issues, and shortages of skilled workers. For cooling systems, a contingency plan will identify the temperature when the existing infrastructure will reach its limit so it is already known when temporary equipment is required. They also outline the level of supplementary cooling equipment required to keep optimal conditions in the server rooms – and determine whether additional power supply is required.
Further to this, contingency plans outline how the equipment will be deployed on site (including any considerations for any obstacles that could prevent this) and where the equipment can be connected to the existing infrastructure. Companies such as ourselves can send experts to the facility to work with the operators to work through the contingency plan, decide the equipment specification and identify the action plan should temporary cooling be required.
Given the fact that hot weather is often periodic, it may not be necessary for operators to invest in permanent cooling infrastructure, especially when consumption is under the microscope. For this reason, temporary cooling systems are a more practical way of mitigating heat, only needing to be integrated when the hot weather hits.
The quick mitigation of heat, however, is predicated on the ability to integrate the specified level of equipment with the existing infrastructure easily. For contractors and consultants, integrating accessible chiller stab-in points, which are tap ins for supplementary equipment to be attached to the cooling system in a data centre, is vital to this.
In the best-case scenario, these can be implemented for a low cost during the facility’s construction phase. However, they can be retrofitted by drilling into the pipe and creating a tap off, however, this is a more expensive procedure. It also risks debris entering and circulating through the system. Putting this action in place during the construction stage reduces the cost and risk for operators later on, so represents a far safer and more reliable choice.
If connections are not already in place, there could be a longer period to install an alternative solution in the server rooms themselves. These could also potentially be less efficient means of cooling and take up excess space inside the facility. However, if this is identified early on in the planning stage, suitable and efficient equipment and where it should be implemented can be decided early on to avoid delays.
For consultants and contractors, being able to make recommendations on ways to alleviate the effects of sudden heatwaves or extreme weather fronts is important. By advising and implementing methods of integrating supplementary cooling early on, building services professionals can help in avoiding downtime and added cost for their clients. Being sure of the facility’s demand, the weather patterns in its location and, crucially, consulting an expert of data centre temperature control will ensure such issues do not occur.
For more information on how Aggreko can help you keep temperature and moisture under control in your data centre click below.