Last week I wrote several predictions for the coming 12 months in the digital media industry. There I wrote that, over time, the way we consume content will change, as we become more and more dependent on the internet.
What I didn’t develop was the fact that this will result in massively increased pressure upon network infrastructure, and this will need to be addressed urgently
Recent research by Ofcom highlighted that, on average, adults watch a whopping 244 minutes of TV a day but only consumed 26 minutes of media via tablets. If just a small percentage of those 244 minutes moved to IP delivery, whether through the TV, a mobile device or laptop, that would put a lot of extra pressure on networks. And we must also add to this pressure the increasing trend for enterprises moving applications to the cloud, 4k/HDR/8k content and the emergence of the Internet of Things/wearables.
If we consider a slightlymore mature market in the US, 34.9% of peak time downstream internet traffic is due to one company; Netflix. Currently Netflix is responsible for 2.5x the traffic that YouTube is responsible for, which is quite startling considering that this market in itself is still in a growth phase.
The Internet today is, to all intents and purposes, a video network. However it was never designed with that in mind. The pressures facing internet connectivity are in contrast to the highly reliable, high quality, never fail broadcast TV service that we as consumers have been receiving for the last 30 years.
At the moment we are reasonably tolerant (excusing the odd rant at the router..) of bandwidth and buffering issues when it comes to websites, but when we’re watching our favourite boxset over IP and it stutters and buffers, are we quite so tolerant? Network optimisation, peering and compression improvements such as HEVC will all help but the infrastructure as a whole needs to be looked at.
Ultimately in the coming year or two we may see the rise of service outages as the network faces considerable pressure. Hopefully this won’t occur, but the way we are consuming content today means that we need to pay increasing attention to the entire connectivity infrastructure, from the enterprises (and where and how they deploy their applications) to the networks (including how they are funded), to the CDNs, the data centre and the router inside our homes.
The Internet was never designed to support video and TV distribution and yet this is increasingly a video network. We are entering an always-on IP, content-heavy world and we need to start considering very seriously how we address connectivity and infrastructure in 2015 or as consumers we might have to realign our expectations somewhat.