*By Juan Fernandes
Disruptive technologies need other technologies or solutions to help them perform the wonders they promise. Some IoT use cases are only possible, for example using 5G, which in turn cannot operate on its own and needs other "agents" such as Multi-access edge computing (MEC).
To better describe the situation, I'll give you some examples, which I'm sure you've heard of. smart manufacturing and smart cities. These two concepts use a set of advanced technologies, such as robotics, artificial intelligence (AI), augmented reality (AR), Internet of things (IoT) and cloud, to obtain technological solutions that provide greater optimization of costs, processes, security, reduction of waste, among other benefits.
In this way, where previously only experienced operators could tell when a machine needed attention or repair, IoT sensors in Industry 4.0 can signal maintenance requirements before they fail, reducing downtime. Basically, data from the Internet of Things feeds AI and machine learning programs so that crucial decisions can be executed.
When we talk about smart cities, we have an urban area that uses technologies and networks to help improve local management of services and infrastructure, reducing operating costs and providing better and faster services and information sharing. And as urbanization accelerates, smart cities are likely to take on even more relevance – according to a Technavio report, this market is expected to move around $2.1 trillion by 2024.
In both examples, from smart manufacturing and smart cities, tons of data needs to be collected from sensors and video cameras, and processed by cloud applications and services before a decision or action takes place. And this is where latency becomes very important and MEC plays its part.
Multi-access computing brings technological resources closer to the end user. Data is processed and stored at the edge of the network, not some distant data center, significantly reducing latency. In this way, MEC provides both an IT services environment and cloud computing capabilities to facilitate the operation of the company, in real time.
So 5G, with its potential for faster speeds, lower latency and higher bandwidth, and MEC can together help bring data processing closer to the end user.
If we go back to the example of the smart city, with the potential for lower latency and to connect more devices per square meter, 5G and MEC could help support the proliferation and scalability of sensors and data collection in an urban environment, thus helping to reduce costs operational. Furthermore, the potential for lower latency and the advent of the private network – to prioritize and segment certain traffic flows over others – can allow for completely “new” use cases, such as connected cars and self-driving vehicles.
But then, do you need a 5G architecture for MEC? The answer is simple: it is not necessary. While 5G services will eventually require high-end computing capabilities with multiple accesses, you don't need the 5G architecture for the multi-access edge computing. In fact, it's possible to deploy mobile edge computing on a 4G network and use it as a launch point for 5G if you decide to migrate later.
MEC gives Communication Service Providers (CSP) and enterprises the ability to host applications and keep data on-premises to minimize latency and maximize efficiency. The solution can be deployed and used at the edge of the mobile operator's network and deployments within customer premises managed or hosted by operators or by cloud providers such as Cloud Edge from AWS, Microsoft Azure, or Google Cloud.
Thus, CSPs don't have to wait for 5G to start competing for 5G services that require large amounts of bandwidth, low latency and localized security. They can start at the edge and prepare for the next step.
* Juan Fernandes, Technology Manager at Logicalis
Notice: The opinion presented in this article is the responsibility of its author and not of ABES - Brazilian Association of Software Companies
EN 
PT 
