Energy Efficiency - 26 billion connected devices across the world, are the challenges set to get worse?
As the need for data centre services continues to increase, aren’t the challenges about energy efficiency only set to get worse?
The continued growth of the Internet of things is set to cause a three-fold increase in IP traffic across the world over the next four years, according to research conducted by Cisco. It predicts that a surge in connected personal devices around the world will mean that, by 2020, there will be 26 billion connected devices across the world, up from 16 billion connections in 2015. The demands on data centres are set to increase significantly and energy efficiency will obviously come into sharp focus. However, it is important to note that the energy efficiency challenges are driving the R&D development of new technologies.
These challenges are pushing innovation forward at a rapid pace in the sector and we are seeing the emergence of more efficient, more cost-effective and more environmentally-friendly solutions to meet these challenges.
Data centres are focused on reducing their operating expenditure (opex) and this is where the latest cooling technology can make a valuable contribution. Compared to compressor technology that is 15 years old, a reduction of at least 35% in energy consumption can be achieved, today. If you replace mechanical cooling with solutions such as free cooling options, the savings can be much greater (up to 65%).
Using the energy efficiency ratio (EER) – which is the ratio of cooling capacity to power consumption
– e.g. 15 years ago the value for typical aircooled chiller was around 2.5-2.7 which is nowadays 3.2-3.7 due to new technologies and/or design concepts. Ideally, the value should be as high as possible (i.e.
not much energy is required to produce the desired cooling capacity). Modern cooling technologies are using significantly less energy compared to 15 years ago, therefore, and manufacturers, such as Stulz, will continue to drive this trend forward through R&D of new, energy efficient technologies.
What are the solutions data centre operators need in order to stay on top of such challenges?
A good solution that is not used properly will lose efficiency. It is just as important to learn how to use equipment, and how to control and monitor the performance, to get the best possible result. As well
as specifying ‘energy efficient’ equipment, staying on top the energy efficiency challenge requires careful consideration of how systems are maintained and serviced. In addition, it is important that solutions are also designed to be environmentally friendly, not just energy efficient, and this needs to be a key aspect of the technology’s design.
What are the pitfalls to avoid when it comes to choosing solutions?
Any project will inevitably involve different parties – including contractors, manufacturers, installers and users – each with different targets. The success of a project will depend on bringing these together and focusing on the overall picture, to get the best benefit from the technology. It is important not to simply focus on one’s own vision of the project. Networking and connecting with these different interested parties, and sharing information and perspectives, is invaluable.
What metrics should be used to measure and monitor power and energy efficiencies – is PUE enough?
In the data centre world, PUE is the most commonly used and established metric. However, I have come from a parallel industry and believe that we should look at more indices. It is important to understand what is the best possible efficiency, or price equivalent efficiency (PEE), achievable in a specific geographical location or region. PUE, as it stands, does not offer a true picture of the efficiency of a data centre on its own. Energy efficiency needs to be understood within the context of the best possible performance that can be achieved in any given location. A data centre in the North is going to encounter very different weather conditions to a data centre in the South, so the possibilities in terms of energy efficiency will be greater. This must be taken into account, to fully understand and measure performance.
The problem is, that the commonly used ESEER figure is related to HVAC for commercial buildings can not be transferred to datacenter applications.
Other reference figures like PUE are just simply overdriving very important factors when it comes to energy efficient design for IT-cooling, such as:
• Location of the datacentre
• Building structure
• CHW system design (glycol or glycolfree)
• Load profiles
• CHW temperature levels
As the value of cooling is usually very important for PUE calculations, a lot of explanations are necessary to be made to really combine PUE figures to each other. The easiest, most transparent and valueable way to compare energy efficiency of cooling systems is a project based energy consumption evaluation.
This approach is virtually displaying the thermodynamical situation on site reflecting all external and internal influences.
Any other points?
Another significant challenge in the short-term will be cyber security. While the energy efficiency of the technology is a key consideration, it will also need to be viewed in terms of being a potential gateway for cyber threats. A lot of information and control is cloud-based, so we will need to focus more and more on addressing cyber security. If a cyber attack targets a specific node point, via the building management system, for example, it is feasible that this could result in failure. The cooling system is an important part of the facility management. Therefore, a solution must be provided to guarantee security on site – this will need to be considered as part of the cooling plans. In the future, there will need to be greater awareness of the potential threat, in the industry, to protect the integrity and security of cooling operations and other critical infrastructure. This is set to become a big challenge for data centres.