By John Hall, Managing Director, Proximity Data Centres
Edge computing pre-supposes ‘blink of an eye’ response times, which in turn, requires low latency networks for transmitting and processing large volumes of data and images as fast as possible. This puts huge emphasis on data centres being physically close enough to the network edge and their connection to local fibre and or 5G/6G networks.
The good news is the UK’s regional fibre infrastructure is advancing at pace through the work of the major fibre providers, carriers and metro network specialists. Connecting to more strategically located edge data centres across the UK, this is ensuring much greater accessibility to high-capacity fibre and greater bandwidth availability, including, in traditionally underserved areas. Moreover, 5G is finally here and 6G is on the horizon.
However, there’s still a roadblock on the full potential of edge computing when it comes to delivering optimised latency and data transit performance, in any UK location. And that is the overly centralised internet exchange model which has been in place since the mid-90s.
Simply put, there is already too much data going through a limited number of internet exchanges. And this is impacting on response times and data transit costs as data volumes grow exponentially. While the traditional centralised model of internet exchanges in the UK has served us well to date and port sizes have increased from 10Mb to 400Gb to help manage surging data traffic, it is no longer viable for all internet traffic to be routed via a handful of exchanges. In the same way that the early ISPs realised that keeping traffic local drove down costs and improved performance, we now need more regional internet exchanges complementing the existing infrastructure model.
Together with edge data centres, regional exchanges play a key role by keeping traffic local, thereby helping ensure the lowest possible response times, while at the same time, preventing user data transit costs from rocketing. The availability of these resources at a local level will also support digital businesses, regional tech hubs and bring significant economic growth to the regions. These will in turn bring new opportunities to the benefit of local communities.
Regional edge data centres are ideally placed to help accelerate the rollout of local internet exchanges and which will support the existing exchange infrastructure. This is why Proximity is in the process of equipping all its current and future regional edge data centres with internet exchange facilities.
An approach such as this allows more data and applications to be deployed closer to the users and devices that really need them. Be this for ensuring a totally immersive experience when playing computer games, navigating the metaverse or performing remote surgery. Not to mention manufacturers and other industrial users who can expect major efficiencies and competitive advantage from acting in actual or near real-time from the vast amounts of data harvested from super-connected plant and production lines, all while leveraging AI technologies such as machine vision and learning.
Equally, in the rush to get data and applications closer to users and customers, it is also vital not to overlook power availability as the other edge data centre prerequisite. A continuing shift to decentralised edge computing is seeing more hybrid cloud and HPC deployment, both of which already have high density power to rack requirements.
An immersive technology application, for example, such as a virtual 3D virtual reality modelling simulation will demand considerable power to rack to support it – it is already seeing as much as 25MW per rack. There is also the bespoke cooling requirements of many of the emerging IoT and AI-based edge applications to consider.