Data centres and the vast interconnecting networks that link them are now absolutely critical to the functioning of our world. They have become the invisible glue that holds our society together during the toughest of times.
The demand for data connectivity during the current crisis has been unprecedented. For example, Microsoft has reported an increase of 12 million users on its Microsoft Teams collaboration and communication online platform, and one VPN provider – NordVPN – has seen a surge of 165 per cent in users of its remote working technology.
As the result of this increased demand, data centres have been classed as ‘essential infrastructure’ across the globe, and their staff as vital to their continuing functioning. Luckily, data centres are geared towards connectivity, so the ability of staff to work remotely, while still keeping the data centre operating as normal, has been very achievable – as we have seen in Italy with the country’s largest data centre, Aruba S.P.A, remaining operational despite being in the epicentre of the country’s COVID-19 outbreak.
A data centre is a building where computer and networking systems are housed for the purpose of storing and distributing data. You could think of them as ‘where the cloud meets the ground. With our increasing use of digital devices, from iPhones and laptops to autonomous vehicles and smart building technology, data centres are the beating heart that enables our interconnected, digital world to function.
As a society, we depend heavily on the reliability of our digital services. In fact, there is an expectation that they will always work, no matter the circumstances.
We have seen how IT outages affect businesses and consumers. With the influx of additional users during the current crisis, Microsoft Teams experienced a more than two hour outage that frustrated thousands of workers as they logged onto the service, firstly across Europe, then the rest of the world. This has served to demonstrate just how dependent we are on this essential infrastructure and how we really can’t function effectively without it. Going forward, it looks like this will become even more important as we transform into an even more connected world, in every aspect of our lives.
It’s a truism that we learn the most valuable lessons during the toughest of times, and when we look back at other disasters there are takeaways that we can learn every time. I’ll never forget the stories of facility managers in New York during Hurricane Sandy in 2012. After the basements flooded, the huge fuel stores that they maintained in the basement became inaccessible as the diesel pumps that normally delivered the critical fuel up to the generators were underwater. Rather than letting their data centre fail, they ferried jerrycans 17 storeys up the building fire stairs by hand. A monumental effort that kept services online at a critical time.
I find myself wondering what lessons we will learn from this crisis? Currently, data centre operators such as NEXTDC, Global Switch and others, are maintaining 24/7 operations and taking precautions with their staff and customers to ensure their health and safety. However, how will the need to quarantine people, maintain social distance, and in many cases, social isolation and lockdown, impact how data centres are designed, built, operated and maintained in the future?
These are not short-term challenges but will extend over many months. Will the inability to cross country and state borders change how we evaluate our supply chains and the technology selections that we make? Will we prioritise price and efficiency, or will availability of a deep pool of local maintainers become our highest priority?
One unique event that has been driven by this medical emergency is the locking down of facilities to anyone other than core staff to protect facility teams from infection. While locking people out is a sensible precaution, it is very hard to continue with normal operations and to fix problems. Finding ways to achieve normal operation will be critical, and some co-location providers are already offering remote tech services, such as Equinix’s Smart Hands, where clients can guide a pair of competent hands remotely to carry out work.
Our facility managers, construction managers and critical system maintainers will have some great lessons to share once we emerge from this. We should take the time to listen carefully to them and find out their pain points, then share them as a learning for everyone.
The next question we should be asking ourselves is whether we need to learn all of our lessons the hard way? Could we start to plan for future disasters now? While many would say that this pandemic is unprecedented, many of the outcomes are predictable and could be readily identified under good risk management planning. With a little imagination, and by considering other disasters.
So far in 2020, we’ve already experienced devastation from bushfires and floods in Australia, a cyclone in Fiji, volcanic eruption in Indonesia and floods in NZ, so we could easily identify different scenarios that would impact our operations, and plan for them now. Many precautions would overlap with normal good practice for a mission critical facility, but a few would be unique and may make all the difference the next time we face an unprecedented challenge of this scale.
I am completely convinced that our industry is critical to our society, however it’s already time to consider where we can improve. The early identification of risks and hazards and then the sharing of problems and failures is going to become essential to us as an industry, as losing the services that we offer will soon be intolerable to the millions of people that rely on us to keep them connected. We are the invisible glue and we should be building unbreakable bonds around our world.
This thinking paper is part of a collection of insights and expertise from Aurecon as it explores leading through and beyond the COVID-19 disruption. Explore our insights here.
Matt Gurr is a Technical Director with Aurecon, specialising in the design and development of data centre facilities, working in mission critical environments with organisations requiring high levels of uptime and reliability.