Magazine Button
How to run increased data centre workloads while delivering significant carbon savings

How to run increased data centre workloads while delivering significant carbon savings

Data CentresDCA FeatureFacilities & ServersInsightsOperations & SystemsPower & CoolingSoftwareThought LeadershipTop Stories

With organisations in the data centre sector expected to deliver on net zero carbon, they must make best use of the tools available to them in order to unlock significant data centre carbon reduction that the wider business requires. Dean Boyle, CEO, EkkoSense, draws upon some of the ways data centre managers can make smarter performance choices.

Today’s data centres have never been busier, with analyst projections suggesting that workload levels are only going to increase over the next few years – with some projecting a 21% CAGR growth rate through to 2025. Organisations are working hard to transform operations and accelerate their processes so that they can execute digitally at scale. And, whether it’s online retail, e-banking, mobile services delivered by 5G or new digital media, there’s an associated IT workload that’s driving further demand for data centre services.

Data centres are also coming under increased pressure to reduce their energy consumption – particularly as the reality of corporate net zero commitments start to bite. With both public sector bodies and private companies making dramatic net zero carbon commitments, there’s going to be sustained pressure on the high energy users within organisations to start making serious carbon reductions.

This places data centres – already established as one of the world’s highest collective consumers of energy – firmly in the spotlight. With the upcoming COP26 UN Climate Change Conference being held in Glasgow this November, there’s going to be an increased focus on securing global net zero and the requirement to reduce data centre carbon emissions will become more intense. That’s why it’s important that data centre IT operations teams are seen to be doing everything they can to deliver quick carbon reduction wins.

It’s perhaps not surprising that many data centre teams find this challenge daunting, particularly as increasing workloads while securing carbon reductions can seem like conflicting goals. The good news is that this can be achieved. Earlier this year, EkkoSense conducted research that analysed the actual cooling performance within live data centres for many of the world’s leading brands. We assessed cooling performance across a sample of some 133 data centre halls and tracked some 33,000 IT racks.

The results found that the current average data centre cooling utilisation level is only around 40% – meaning that most organisations are not even using over half their cooling capacity. Our analysis also showed that operations are missing out on proven ways of cutting cooling energy consumption by up to 30% using the latest software-based optimisation approaches.

Organisations clearly want to deliver on carbon obligations, but that can be challenging when data centre teams don’t always have a clear understanding of exactly how their rooms are performing from a cooling, capacity and power perspective. Indeed, when faced with an external issue – such as an increased thermal demand placed on facilities by a surge in hosted services – the default position for many operations teams remains to just keep throwing more cooling at the problem. This simply adds to the data centre’s overall carbon footprint and often does little to resolve the original issue.

Data centre operators also need to recognise that optimising thermal performance positively impacts data centre risk management – however, it’s difficult to ask the right questions if you don’t actually have any granular visibility into how individual racks and cooling equipment are performing. Our research showed that only 5% of data centre M&E teams currently monitor and report equipment temperature actively on an individual rack-by-rack basis – and even less collect real-time cooling duty information or conduct any formal cooling resilience tests.

So how should data centre teams optimise their performance?

While operators remain keen to secure carbon reductions, the reality for many is that they simply don’t have access to the tools that can help them to make smart data centre performance choices in real time. And, while legacy DCIM tools might be useful at helping operations teams manage their facilities, many find them limited when it comes to the deep data analysis needed to really optimise performance at the mechanical and electrical level.

So perhaps it’s time to stop treating efficient data centre operations as a black art. You don’t need over-complex DCIM suites or expensive, non-real-time and often imprecise external CFD consultancy to tell you what’s going on across your own data centres. It’s much more useful to have an approach that gives you a real-time dynamic viewpoint of your mission-critical estate.

That’s why, for true data centre infrastructure management, M&E reporting tools need to get much more granular. The good news is that with the latest generation of software-driven data centre optimisation solutions, there’s a real opportunity to achieve this and – in turn – start to unlock the significant data centre carbon reduction that the wider organisation needs to deliver.

In contrast to traditional DCIM solutions that can take years to implement, software-driven thermal optimisation gives data centre teams much faster access to the insights they need for less cost and less human management overhead. At EkkoSense, we have found that those organisations which deploy software-based optimisation solutions for critical infrastructure are able to reduce their cooling energy usage by up to 30%, while at the same time releasing cooling capacity and unlocking immediate carbon savings.

Key innovations that have enabled this to happen include:

  • The application of Machine Learning analytics built right into the heart of the solution, drawing not just on PhD-level thermal expertise but also data from 50m+ data points in critical facilities around the world.
  • Truly granular levels of sensing – taking advantage of the latest low-cost IoT wireless sensor technology to allow sensors to be deployed in higher numbers across the data centre right down to rack level – making true Machine Learning-based analytics and real-time thermal management of critical facilities a reality. This is typically complemented by vendor-agnostic cooling units that provide real-time cooling duty information.
  • Use of the latest web technologies – including gaming interfaces, to provide data centre teams with the most intuitive, easy-to-use and simplest to manage monitoring and management capabilities. By creating immersive Digital Twin representations of your data centre, operations teams get to see all their current cooling, power and thermal conditions via a single, accessible 3D visualisation.
  • Lightweight and easy-to-manage M&E capacity management – meaning that you don’t need an army of people or huge costs to deliver an effective centralised capacity management process.
  • Having 24x7x365 access to your own virtual PhD expert who’s always busy continually optimising your facility and always on hand to help in-house data centre teams to deliver the next best optimisation outcome.
  • Unique Cooling Zones functionality that tracks real-time correlation between cooling units and IT racks to support optimisation and resiliency activities.

To date, most of the traditional solutions classed as DCIM have only really offered very basic M&E monitoring capabilities rather than the truly integrated infrastructure management offering, initially promised by vendors. However, DCIM doesn’t need to be that way in 2021. It’s now possible to deploy software-based optimisation solutions that not only pick up on potential data centre problems or underlying negative trends, but also suggest recommended actions to resolve thermal issues, simplify capacity management and optimise cooling energy. It’s DCIM that works, without all the hassle and at a fraction of the cost of traditional DCIM systems.

Click below to share this article

Browse our latest issue

Intelligent Data Centres

View Magazine Archive