Operators are getting closer to the edge -- the network edge, that is. Mobile edge computing, or multi-access edge computing as it's now being referred to, has finally made the leap from concept to reality. It was inevitable, especially with 5G on the horizon and the anticipated surge in IoT traffic coming down the line. Moving cloud, compute and processing power to the network edge makes it possible to support the ultra-low latency, interactive requirements of 5G applications (such as connected cars) that run at the edge of mobile networks.
This much we know, but what we don't know is exactly how those 5G applications are going to behave and interact with the network. The other grey area is IoT, because it's still not clear what type of traffic patterns will emerge once networks have to contend with fleets of driverless cars, automated factories and connected production lines. Operators could find it a bit tricky if they don't have an accurate view of what's happening across the 5G network, especially when they're dealing with a complex new infrastructure.
But hold on, we've been here before. Surely, this is just a simple transition from one wireless technology standard to another? Well, not exactly -- the fundamental issue operators face is that IoT and 5G is completely unprecedented. There is no blueprint for the IoT and at the moment the only data available is based mostly on predictions or careful modelling, and herein lies the problem. Operators are investing in new equipment, but they're basing future scenarios and capacity demand on current data traffic trends and volumes. However, this approach isn't sustainable, and monitoring IoT traffic as it begins to increase and fluctuate will become a full-time job for operators. They will find themselves constantly having to adjust network conditions just to maintain optimal levels of performance.
However, by effectively moving data center capabilities to the edge of the network, operators will bring new levels of sophistication to their infrastructure. The smart data this will generate will add much needed intelligence to the network, allowing operators to allocate capacity to areas where it's needed the most to manage IoT traffic. It will also let operators know if there's a problem with their 5G network before it escalates and hundreds and thousands of IoT devices stop communicating with each other.
Put it this way: A frustrated subscriber who can't stream the latest Game of Thrones episode will call the customer service center to let you know there's a problem, but machines on a production line that are having connectivity issues won't be flagging that to anyone. So how do you find out if there is a problem with your IoT network?
The only way to gain insight is to leverage smart data solutions that maintain visibility throughout the IoT lifecycle. This covers everything from the testing phase through to the monitoring and analysis of live traffic, to network orchestration and automation. In fact, smart data is absolutely key to network automation and when used in conjunction with policy, it will enable operators to tackle traffic congestion and data throughput issues. Real-time smart data can be fed into the network policy engine (for example, the PCRF in 4G) to drive decision making and resource allocation -- spinning network resources up and down as needed to deal with live conditions.
This type of automated response will be essential for managing unpredictable IoT traffic. It will also help to reduce capex by removing the need to install new radio equipment to beef up cell sites. Instead, operators will be able to make intelligent use of their existing network infrastructure.
Multi-access edge computing is a nascent space, but it will become a critical element of mobile network architecture. However, this isn't just a mobile play -- it will also apply to ISPs, cable companies and other service providers as 5G and IoT become more pervasive. So the need for visibility across all phases of the IoT life-cycle are abundantly clear as the industry moves network infrastructure to the edge, harnessing NFV and cloud to deliver new services. With a smart data solution, operators can monitor the sheer depth and breadth of the IoT ecosystem to ensure everything is assured and connected.
— John English, Senior Manager, Service Provider Solutions, NetScout Systems