Time to call out the mission-critical data centre capacity myth?

04 May 2021

Martin Docherty, Ekkosense

I have often thought that there’s a lot that data centre operations teams and facilities management could learn from their colleagues in the hospitality sector. It’s not difficult to analyse a hotel's occupancy rate - the rooms are either empty or occupied! Without the right toolset, it is slightly more challenging for data centres to check their occupancy.

So it’s perhaps unsurprising that DC ops teams spend so much time worrying that their data centres are fully occupied and maxed out with no space left for additional racks. This perspective is largely undone by the latest EkkoSense research that clearly shows that this isn’t - and probably never has been - the case. In fact, our in-depth analysis found that, on average, data centres still only utilise around 40% of their expensive cooling capacity.

Next time you think there’s no spare data centre capacity, or you’re contemplating further capex investment to support additional workloads, ask yourself, just how much of your data centre estate capacity are you really using? This question is often surprisingly hard for data centre or FM teams to answer. Even the best operations professionals can find this hard to provide a current, accurate response, particularly as operators don’t always have a clear understanding of how their rooms are performing from a cooling, capacity and power perspective.

Gaining a real-time, dynamic viewpoint of what’s going on

Organisations simply can’t afford to keep on treating efficient data centre operations as a black art. The reality is 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 in your own data centre. It’s much more valuable to have a real-time dynamic viewpoint of your mission critical estate.

For true data centre infrastructure management, however, M&E reporting tools have to start getting much more granular – drawing on the latest low-cost data centre IoT sensor technologies and intuitive 3D software visualisations to show rooms in a realistic 360° real-time digital-twin view. This makes the immersive real-time optimisation of data centers a reality for operations/facility teams. By gathering and visualising this data at a granular level, they can start to identify how individual racks and cooling equipment are currently performing.

However, it’s not enough just to monitor performance. With the latest generation of software-based data centre optimisation tools, teams can start to draw on powerful machine learning and AI-based analytics capabilities to secure actionable improvements as part of their efforts to actively manage and maximise the performance of their critical data centre environments. Such an AI and machine learning-led approach doesn’t just highlight problems as they occur but can now actually deliver intelligent insights and recommended solutions before the issues even develop.

Removing barriers to data centre effectiveness

One of the main barriers to data centre optimisation has always been the sheer complexity of traditional DCIM tools and CFD solutions. At EkkoSense we believe that if you want people to make the most of your software you have to make it as easy as possible to use. That’s why we’ve drawn on powerful 3D visualisations and gaming software techniques to make our solution much more accessible.

For example, having a 3D visualisation approach and a simple drag and drop interface to support a range of M&E capacity planning activity - from basic rack changes through to complete new room layouts - makes day-to-day operational data centre management so much easier. Similarly, capabilities such as space planning and reserved space allocation can help organisations to unlock stranded capacity from their existing data centre cooling and power infrastructure – effectively enabling them to do much more with less!

Adopting this approach can also be used to manage allocated and reserved power, monitor rack power usage, and also provide comprehensive ticketing and future capacity support by taking advantage of Live Views to reserve space, power and cooling for future projects. And with effective thermal optimisation lowering energy costs and releasing further capacity, the good news Is that software-based data centre optimisation can deliver payback in under a year with benefits available immediately.

With the digital transformation agenda demanding that more and more interactions go through data centre and switch sites, it’s re-assuring for operations teams to know that they may already have the capacity in place to scale up their IT activities. Immediate benefits will include potentially higher revenues for co-location sites, or being able to get more out of their existing cooling assets – effectively delaying the need for further capex investments.