Multi-access edge computing (MEC): the whats, whys and hows

In traditional cloud computing, your software is many miles from the server that makes it run. The journey from device to backend can be many thousands of miles. 

This might not seem that odd to you. After all, we tend not to have power supply infrastructure in our back gardens. We're used to resources travelling long distances to reach us. 

But with cloud computing, the weird thing is that the user and backend are in constant dialogue. If you make a request on a cloud app, the backend has to formulate a response. That data has to go all the way around the world only to end up back where it started. 

It's a data boomerang that's the equivalent of a skyscraper with no phones. Every time the receptionist wants to speak to someone on the top floor, they have to run up hundreds of stairs and back. 

Edge computing is the answer to this problem. It brings compute capacity closer to the user and their device. It does this by installing devices on the network itself, whether at a base station or central office. 

If this is new to you, we regret to tell you that it's already getting superseded. Edge computing is fast becoming replaced by multi-access edge computing or MEC (pronounced to rhyme with "dreck"). 

In this article, we take a whistle-stop tour of MEC's capabilities and benefits. But first, a note on terminology. 

The ABC of MEC 

The cloud computing industry is a hotbed of ingenuity. It's not, however, a hotbed of wordsmiths. 

Anyone who's dipped their toes into the cloud world will have found their feet in a pool of alphabet soup. Here at Ascend Cloud Solutions, we're keen to make this lingo as easily accessible as your Netflix account. 

MEC, then. When the term was coined, it stood for "mobile edge computing". But times have changed, and it's now (mostly) used to mean "multi-access edge computing". 

This is because edge computing is no longer the preserve of mobile networks. It also applies to fixed networks, converged networks and even WiFi. 

MEC solutions are associated with the European Telecoms Standard Institute (ETSI). This organisation seeks to standardise edge computing to make devices and solutions interoperable across the telecom industry.

ETSI's standards are constantly evolving as it pursues its mission of one common platform, interface and set of APIs for developers in all industries. 

A simple solution to a complex problem 

Networks are more complex than ever. The global network was already starting to sprawl when COVID-19 made hybrid working the new normal. 

Today, a network isn't just a load of computers hooked up to an office server. You could have a Teams meeting where one participant is at central office, one is on a train and one is at a conference in Chibougamau. 

Traditional networks don't cut it anymore. MEC is an intuitive, easy-to-deploy solution that brings compute capacity closer to the user, whether they're in Azerbaijan, the Azores or Aldi. 

How does edge computing work? 

Edge computing isn't rocket science, but it's not bottle-rocket building either. To go deep into the technical detail is beyond the scope of this article, so let's explore a relatively simple example. 

In a traditional cloud model, the user makes a request on an app. This gets sent to the backend server via the core network. A response is formulated and sent back to the end user. 

This is all well and good. But the distance travelled means that latency is a given – especially when using data-heavy applications such as HD video. 

Today, people increasingly expect real-time responsiveness. At times, traditional cloud computing can struggle to deliver this. 

Edge computing installs a server closer to the data, reducing latency and the volume of data that's being sent across the network. 

So much for edge computing. What difference does "multi-access" make? 

Multi-access edge computing: an example 

Edge computing installs a server on the network itself, lowering latency and the volume of data transfer. MEC adds a third element: mobility. 

Edge computing is built for fixed locations – image analysis software at a factory, say. But in the age of hybrid working, mobility has to be a priority. A scattered workforce needs a responsive service no matter where they are.

 MEC, then, is where multiple edge servers are installed to boost accessibility and workforce mobility. 

There are two major benefits to this. First, it reduces latency even more than simple edge computing. 

Secondly, it reduces the load on the backend server. These savings can be repurposed into machine learning training. If an app is running video analysis, for instance, MEC takes some of the weight off the backend, freeing up capacity that can now be used to improve the quality of the analysis. 

How MEC can take the sting out of downtime 

The case for MEC is relatively simple: it brings capacity closer to the user, meaning a faster, more efficient and more responsive service. 

But it also means outages are less critical than with traditional cloud computing models. Operations can continue even when power is down between the backend server and the app. 

What are some of MEC's applications? 

MEC can be used wherever there's a need for additional compute capacity. A factory, for instance, could use MEC to process confidential data on-site. This means lower costs as well as greater efficiency. 

In the gaming world, MEC would mean an even faster response time. That means you can explore virtual worlds without glitches, lags or other nuisances. 

Finally, MEC makes smart cities possible. If vehicles are transmitting information to each other, say, they need to be able to do so quickly and on the move. 

The bottom line 

MEC is all about bringing compute capacity closer to the user. And at a time when networks are more complex than ever, this is just what the cloud doctor ordered. 

Ascend Cloud Solutions is a team of VMware migration specialists with decades of experience and a global client base. Need help navigating the cloud? Get in touchtoday for a quick, competitive quote.