Solving 2022’s key data centre optimisation challenge

The challenges of running a data centre are ever increasing, as is the demand for running carbon neutral facilities. Here, EkkoSense identifies the five issues operations teams will need to resolve to successfully balance increased workloads while still securing carbon savings.

EkkoSense has identified five key issues that data centre teams need to address to solve 2022’s key data centre optimisation challenge. Faced with escalating IT workloads, operations teams are also increasingly being asked to reduce their energy usage – two potentially conflicting goals.

“Currently, very few operations have either the real-time data centre cooling, power and capacity performance visibility or available in-house optimisation expertise they need to meet these conflicting goals – increase workload while also reducing energy,” said Dean Boyle, CEO at EkkoSense – a leader in the field of software-driven data centre optimisation solutions. “And with critical facilities now coming under increased pressure to reduce their carbon consumption, particularly as corporate net zero commitments start to bite, 2022 looks like it’s going to be another challenging year for data centre teams.

“However, it’s going to be hard for operations teams to address both data centre workload and energy saving goals unless they prioritise performance optimisation across all their critical facilities,” said Boyle. “And this means knowing exactly where they’re starting from in terms of energy usage, and also in data centre resourcing. Sustained pressure on the high energy users within organisations mean that the days of simply throwing more cooling at workload issues are over. For 2022, the focus will need to be on smarter resourcing, data granularity, visibility, empowering operations with AI, and measurable net zero contributions.”

1. Be smarter with your resourcing – If data centre teams are being tasked with unlocking carbon savings, it’s essential that they have the right people in place to drive energy savings across their operations. But with demand for digital services showing no sign of slowing down, 2022 will continue to see a shortage of skilled professionals across data centres, with recruitment concerns amplified by ‘the great resignation’. This will place an increased emphasis on decoupling skilled staff from specific locations, with the latest cloud-based optimisation solutions helping teams to optimise cooling capacity and minimise energy usage – regardless of location. The impact of the pandemic – particularly the difficulties involved in getting people into a critical facility plus the emergence of remote ‘dark’ Edge sites – means that people will be focused on enhancing remote monitoring for 2022.
2. Improve data granularity – For true data centre performance optimisation, it’s essential to monitor and report temperature and cooling loads much more actively. ASHRAE suggests a minimum of three temperature sensors per rack, but to achieve this probably requires around 10x more sensors than are currently deployed in today’s typical data centre. EkkoSense research shows that only around 5% of M&E teams currently monitor and report actively on an individual rack-by-rack basis – and even less collect real-time cooling duty. However, if you want to track cooling loads in real-time – and generate the kind of Machine Learning insights you’ll need to enable true AI-powered optimisation – there’s no option but to go granular. And with the availability of the latest ultra-low-cost wireless sensors, there’s really no excuse not to do this in 2022.
3. Increase your visibility with digital twins – One of the main barriers to data centre optimisation has always been the complexity of traditional DCIM and CFD solutions. That’s why, if you’re seriously looking to optimise performance, it makes sense to make things as simple and accessible as possible for operations teams. Building an intuitive ‘digital twin’ of your data centre lets you take things to the next level, providing clear visualisations that help to successfully communicate inevitably complex data. This, in turn, makes it much easier to see what’s happening across your estate, highlighting potential anomalies and displaying suggested airflow and cooling improvements.
4. Empower your operations with AI – While the vision of data centre automation remains attractive to CIOs, the reality is that automated controls are much less able to accommodate the continual, inevitable workload changes that always happen in data centres. So, rather than treat Machine Learning and AI as a universal solution, we expect to see data centre teams in 2022 focusing on those areas – such as cooling optimisation and airflow management – where these technologies can be applied and deliver significant results. But rather than rely on unwieldy automation solutions, we see a light touch DCIM approach – with operations teams provided with AI-powered, actionable recommendations – proving a smarter choice in 2022 and beyond.
5. Secure measurable net zero contributions – At COP26, both countries and leading brands signed up to ambitious net zero targets, and 2022 will start to see a much tighter link between financial and carbon budgets. In the UK, proposed Treasury rules suggest that financial institutions and companies with shares listed on the London Stock Exchange will need to come up with net zero transition plans that detail their greenhouse gas emissions targets and the steps they will be taking to achieve them. Unfortunately, it’s difficult for data centre teams to deliver a precise carbon reduction on their operations if they don’t know exactly how much energy they’re using in the first place. Our own research at EkkoSense suggests that very few teams really know how their rooms are performing from a cooling, capacity and power perspective. This will need to change if data centres are to make a significant contribution to corporate carbon reduction activities.

“The good news for data centre teams looking to support both increased workloads while also securing quantifiable energy savings in 2022, is that pursuing a software-enabled approach to data centre performance optimisation can help achieve these goals,” said Boyle. “Combining the collection of granular real-time data, understanding the zones of influence that exist between specific cooling assets and the racks they’re cooling, and unlocking the potential of AI-powered optimisation recommendations will help to put control back in the hands of data centre teams. Our experience suggests that following this kind of approach can help keep data centre teams on track in their journey to achieve cooling energy savings – with average savings of 30% already achievable.”