Mobile edge computing, or multi-access edge computing (MEC), enables cloud computing capabilities and an IT service environment originally at the edge of a network.
MEC is a type of edge computing, a distributed architecture that reduces latency by housing applications, data, and computing resources at locations geographically closer to end users. It expands on cloud computing capabilities by placing those capabilities in the network closer to the end users, reducing mobile network congestion and decreasing latency.
MEC uses multiple aggregation (multi-access) points on the network to reduce the distance that data travels even further.
Conventionally, a client application makes a request to a server. Then, the server processes it and, depending on the type of request, returns the proper response to the client. This process is subject to latency, which is the delay between a user’s action and a web application’s response due to the distance between the client and server.
An edge compute solution is the first step in decreasing latency by minimizing the distance data must travel between users and the resources they are accessing. This directly results in a faster response.
However, using edge computing in this way has some drawbacks. Namely, the edge computing architecture is only helpful for those users near the edge node. For example, if the edge node is in North America, someone from Asia trying to visit the application would not see any improvements.
To prevent this, you must install edge nodes in every region your users are located, which can significantly increase the operating cost. That’s the same challenge that MEC seeks to solve.
The MEC architecture can be divided into three main sections, each outlined below.
The MEC platform manager is responsible for several critical functions, including the communication between application-level events and the application’s lifecycle management.
Since computation, data processing, and resource allocation all happen on the network edge, they require a component to manage them efficiently. This is where the virtualization infrastructure manager (VIM) comes into play.
VIM is a core component of the platform manager responsible for allocating, assigning, and releasing storage or computing resources. To do this, VIM uses a virtual network service called Virtual Network Function (VNF). Each VNF has a virtualized layer that divides the storage, computing, and network resources according to their requirements.
The MEC host’s virtualization infrastructure composes all the required infrastructure to store and process tasks. A MEC host consists of three components: virtualization infrastructure, the MEC platform, and edge applications.
The MEC platform is located within the MEC host and interacts directly with edge applications to handle the following tasks:
Virtualization infrastructure directly communicates with VIM to supply resources to edge applications. MEC architecture contains multiple MEC hosts and MEC platforms.
The MEC Orchestrator is the last crucial piece of the MEC architecture. Its role is to maintain an overall view of resources and services. It can also select the appropriate available host according to the type of application, host latency, and resource requirements.
Because MEC has such a significant architecture, it has operational standards that all parties must meet, including vendors, network providers, and developers.
The ETSI formed the Industry Specification Group (ISG) to create and oversee MEC standards that support the creation and maintenance of an open standard system and application programming interfaces (APIs) using common programming models, relevant tool chains, and software development kits to encourage and expedite the development of new applications for the new MEC environment.
These standards allow different stakeholders to coexist and benefit equally from MEC. The standards can be divided according to the API they facilitate. For example, ETSI has system, host, platform, network, and application management standards.
Even though MEC architecture is recent, several applications have already adopted it. There are several real-world use cases for MEC — some of which are outlined below.
Currently, many auto providers are focused on setting up their cars with connected services. For example, luxury car makers have recently introduced real-time traffic light information. According to the data, it tells the ideal speed to drive to avoid red lights.
MEC facilitates real-time and reliable processing, on which these advanced automotive features depend. Furthermore, many car makers provide real-time traffic, weather, and pedestrian information and then process it to achieve autonomous driving. This system must be real-time as safety and security require split-second decision-making.
The AR and VR industry has taken this world by storm. Many open-world games have extensive data and computing requirements that traditional cloud architecture cannot sustain, leading to poor user experience.
Because MEC is very low latency and includes computing in the network, you can use MEC to offer a smoother and better gaming experience. Furthermore, ultra-low latency makes it possible to play multiplayer games across the regions.
There’s been a sharp increase in the demand for surveillance systems in cities and organizations to mitigate all threats. One of the critical aspects of that is the ability to perform real-time threat detection. But, in the practical world with traditional systems, analyzing thousands to millions of people in real-time for threat detection is impossible.
Because MEC supports low-latency communication and faster computation in the network hub, it’s well-suited to security applications, which require efficient and reliable processing and response time.
Many notable applications in mobile edge computing are becoming multi-access. For example, active device location tracking allows operators to track active terminal equipment independent of Global Positioning System (GPS) location. This is based on third-party geolocation algorithms within an MEC application server’s host.
Another use is distributed content and domain name system (DNS) caching, which reduces server load and speeds up data delivery to customers.
Additionally, MEC is used for content delivery, big data analytics, edge video caching, collaborative computing, smart venues, smart enterprises, and more.
