EDITOR ’ S QUESTION
DEAN BOYLE , CEO , EKKOSENSE
efore focusing on specific data centre technologies
B for 2022 and beyond , it ’ s worth identifying the two major challenges that operations teams face over the coming year .
Firstly , data centre workloads are growing , with analyst projections suggesting that levels could see CAGR growth rates of up to 21 % between now and 2025 . Secondly , organisations are now coming under increased pressure to reduce their energy consumption – particularly as the reality of corporate net zero commitments start to bite .
Unfortunately , these two very real challenges appear conflicting . Meeting projected workloads will be hard to achieve without growing data centre capacity and associated cooling infrastructure . This in turn adds to the data centre ’ s overall carbon footprint – exactly what data centre management doesn ’ t want at a time when they are being tasked with securing quick carbon reduction wins across their operations .
The default position for many data centre teams is still to keep throwing more cooling at a problem should any issues arise . In EkkoSense research conducted earlier this year , we identified a massive industry-wide over-provision of cooling , with average cooling utilisation currently sitting at just 40 %. Despite this excess cooling , some 15 % of racks still remained outside of ASHRAE guidelines for inlet temperatures .
Perhaps the reason for this is that only 5 % of M & E teams currently monitor and report actively on an individual rack-byrack basis , and even less collect realtime cooling duty information .
This means it ’ s very difficult for data centre teams to deliver a precise carbon reduction across their operations if they
WE BELIEVE DATA CENTRES SHOULD NOW BE PRIORITISING A MORE EFFECTIVE SOFTWARE-BASED OPTIMISATION APPROACH .
don ’ t know exactly how much energy they ’ re using in the first place .
That ’ s why we believe data centres should now be prioritising a more effective software-based optimisation approach – one that truly embraces the power of technologies such as IoT devices , Machine Learning and AI . With effective optimisation and ongoing monitoring and management , organisations will be able to increase their data centre cooling utilisation and secure average data centre cooling energy costs of 30 %.
And rather than relying on unwieldy automation solutions , we expect a much more light-touch DCIM-style approach becoming the smarter choice for 2022 and beyond . For this , we see cooling , power and space data being collected at a highly granular level – ideally with each individual rack featuring multiple sensor points . Operations teams can then access 3D visualisations that are easy to interpret , while AI algorithms will draw on potentially billions of Machine Learning data points to provide actionable optimisation insights . The key difference for operations teams will be that they are provided with actionable recommendations that they can validate before pursuing .
Optimisation done this way via a new style of light-touch DCIM will help to put the control back in the hands of data centre teams , giving them the insight they need to make the kind of smart optimisation choices that can actually reduce cooling energy usage across their sites . ◊
Content supplied by the DCA
32 www . intelligentdatacentres . com