Prepare for the autonomous network
With the ever-increasing demand for cloud computing, mobility and video streaming, there is a growing burden being placed on our global networks, writes Loudon Blair, a senior director in the Corporate Strategy Office at Ciena. Indeed, system overload is now an imminent possibility for many organisations due to the vast amount of traffic coming – and yet to come – from Internet of Things’ (IoT) devices, 4K and 8K video streaming and virtual reality applications. This is of course in addition to the substantial amount of over-the-top (OTT) service streaming of voice and video already commonplace in daily use.
Network traffic is forecast to grow 25% annually over the next five years, according to the Ovum Network Traffic Forecast: 2015-20. Network operators will have to cope with traffic surpassing one million petabytes per year by 2020. In order to keep up with these web-scale demands, firms now need to make sure network infrastructure works smarter, not harder.
The current terms we use for optical transmissions rate are in tens and hundreds of gigabits per second – 10G, 100G, 400G. However, with the increase in demand, terabits are about to start to dominate the conversation. Network operators are now facing pressure on their systems and need to scale their networks to respond in real-time to unpredictable traffic demands. However, it isn’t realistic to simply add networking resources in the same way it has been done in the past – cost and capacity limitations have to factor into the equation too. It’s now time to understand how the network’s optical layer, powered by advanced software, can be used to help meet web-scale demands.
Move on from historic network designs
Before now, networks were designed by manually entering structures into a spreadsheet or planning tool, then factoring in extra performance capability – or optical margin – to the original design to be prepared for any system aging or an unusual level of demand. All of this is changing. Moving away from this inflexible approach, future networks will be able to manage their own performance level and automatically self-optimise or adjust so that the needed capacity and load balancing can be achieved at any moment. In much the same way that self-driving cars offer the promise of less congestion and fewer delays, networks will become self-driving to ensure 24/7 uptime. In turn, self-driving, autonomous networks will be able to provide the robust speed and high-quality experience that customers now expect.
New tunable optics and advanced software capabilities are examples of advancements in coherent optical technology which are enabling the network to become much more programmable and agile. At the moment, operators do not have real-time data from the network or the analytic tools needed to make fast and informed decisions. As a result, inaccurate capacity assumptions are being made, leading to sub-optimal use of network resources and reduced profitability. The intelligent network will change all of this. Through network monitoring and information gathering in real time, combined with automated, intelligent decision making, networks will know what current and likely future demand will be, and compensate accordingly.
How the optical layer facilitates the self-driving, autonomous network
The optical layer can enable the autonomous network in two equally important ways:
- Deterministic path flows. It’s no longer enough to simply focus on adding capacity to a set of network links. Whilst having abundant levels of capacity is useful, without the insight into where it should be directed, the likelihood of inefficiency and waste is high. Adding intelligence to the optical layer makes it possible to know how and when to shift capacity around. For example, one route might only give an unprotected connection for basic services whereas an alternative one could support high availability for mission critical services. Understanding these options is important. For example, when the next viral app phenomenon like Pokemon GO is introduced, the network must be able to change quickly so that users do not experience lost service or delays to their enjoyment.
- Upgrade capacity availability in real-time. The ability to move capacity automatically, without delay or manual inputs, is the key to constructing a truly self-driving and autonomous network. With the adoption of 5G, users will be able to access content whenever, wherever, requiring a network that gives the bandwidth they need, when they need it. Whilst operators can plan ahead for large scheduled events, such as global sporting events, and predict capacity as such, it’s the more irregular news events or future video streaming trends that are harder, or even impossible, to forecast.
The other result of maximising resources is being able to take advantage of unused capacity in order to improve cost efficiency and service experience. In the past, systems were set up to operate with a certain fixed data rate over a maximum distance, for example, 100G, over 2,000km. However, in a real network, links can widely vary – they can be 95km or 372km or 1,125km. Due to this, many older systems were designed predominately for those links, wasting network resources. However, with the rise of autonomous networks, intelligent systems can adapt the data rate for the appropriate distance. For example, if data only needs to travel 95km, the rate can be tuned to 400G, increasing available bandwidth without changing hardware. This new network takes advantage of the technical capacity-distance trade-off. Even in the case of a system breakdown, intelligent decision-making can adjust the data rate on the fly, so when the network is forced to reroute along a longer backup path, the network can still operate services, albeit with reduced link capacity – only degrading the service bandwidth without failing completely.
The coming year is the time that the foundations for self-driving, autonomous networks of the future must be laid, in order to support the arrival of next generation network services and the spiraling demand already mentioned. Focus is already shifting to improving the performance of the network with software solutions on top of adding capacity and increasing profit margins. As a result, customers will be able to benefit from seamless connectivity via an automated traffic management system. With capacity deployed where it is needed most, no one should miss out on the latest Netflix series or online game at peak periods.