Multi-access Edge Computing (MEC)
This article explains Multi-access Edge Computing (MEC), how it brings computing resources closer to users and devices within the access network, how MEC works, and why it is useful for applications that require fast and local processing.
1. Why Do We Need MEC?
As discussed in the previous article: ▶ On-Device vs Edge vs Cloud Computing
A mobile device (such as a robot or phone) can send its data to a remote cloud server for processing. However, the cloud server may be located far away from the device.
For example:
Mobile Device
│
▼
Cellular Network
│
▼
Remote Cloud
│
▼
Processing
In this approach, data generated by the mobile device must travel through the mobile network to reach the remote cloud.
General concept is this: The edge-computing concept is to move the computing resources closer to the device.
For example:
Mobile Device
│
▼
Cellular Network
│
▼
Edge Server
│
▼
Processing
This means that the device can use a more powerful computer located near it, rather than performing all of the computation locally or sending everything to a distant cloud.
Specific conept is this: For mobile devices, this edge-computing concept can be implemented as part of the mobile network infrastructure. Telecom operators can deploy computing resources within or near their access networks so that mobile devices can use nearby computing resources.
This is where Multi-access Edge Computing (MEC) comes in (i.e. INSIDE MOBILE NETWORK INFRASTRUCTURE SUCH AS TELCO COMPANIES)
MEC is the implementation of the edge-computing concept within a mobile and access network, where telecom operators provide computing resources close to mobile users and devices as part of the network infrastructure.
For example:
Mobile Device
│
▼
Mobile Tower
│
▼
MEC Server
│
▼
Processing
The MEC server can therefore process data close to where it is generated while being integrated into the mobile network.
This is useful for applications that require significant computing resources but cannot perform all of the processing on the mobile device itself. Examples include augmented reality, video analytics, connected vehicles, industrial applications, and other mobile applications that require computationally intensive processing.
The basic relationship is therefore:
Edge computing is the general concept of bringing computing closer to resource-constrained devices. MEC is this concept implemented within the mobile network by telecom operators.
The next section looks at what Multi-access Edge Computing actually means and where the MEC server is located within the network.
2. Where is the MEC Server Located?
The main idea behind MEC is that the computing resources are placed close to the users and devices. In a mobile network, the MEC server can therefore be placed at or near the network edge, close to where mobile devices connect to the network.
For example, a MEC server may be deployed at a cellular base station or at a location close to the mobile tower.
Mobile Device
│
│
▼
Mobile Tower
│
│
▼
MEC Server
(Nearby)
│
▼
Network
In a larger deployment, the MEC server does not necessarily have to be physically next to the tower. It can be located at another nearby network facility while still remaining close to the users and devices.
For example:
Mobile Devices
│
▼
Mobile Towers
│
▼
Access Network
│
▼
MEC Server
│
▼
Core Network
│
▼
Cloud
The exact location of the MEC server depends on the network architecture and the requirements of the application.
The important point is that the MEC server is placed closer to the access network and its users than a typical remote cloud data center.
This allows mobile devices to access computing resources without sending all of their data to a distant cloud.
In MEC, the edge is not simply a geographical location. It refers to computing resources placed close to the point where users and devices access the network.
This placement is what allows MEC to provide computing resources close to mobile users while still keeping the devices themselves relatively simple and resource-constrained.
3. How Does MEC Work?
The main purpose of doing this is to provide several benefits, including:
- Lower latency because the computing server is physically closer to the device.
- Reduced network traffic because data can be processed near where it is generated.
- Faster response for applications that require immediate results.
- Reduced load on remote cloud infrastructure because some processing is performed at the edge.
For example, consider a mobile device that needs to perform an AI-based image analysis.
Mobile Device
│
│ Image
▼
MEC Server
│
│ AI Processing
▼
Result
│
▼
Mobile Device
The mobile device captures the image and sends it to the MEC server. The MEC server performs the computationally intensive processing and returns the result to the device.
Because the MEC server is physically close to the device, the data does not have to travel to a distant cloud data center for processing. This can significantly reduce the time required for the device to obtain the result.
The MEC server can also provide computing resources to multiple devices connected to the same access network.
Device 1 ──┐
Device 2 ──┤
Device 3 ──┼──► MEC Server
Device 4 ──┤
Device 5 ──┘
For example, several cameras, robots, or mobile devices in the same area can use the computing resources of a MEC server instead of each device having to perform the computation independently.
Therefore, MEC allows resource-constrained devices to use more powerful computing resources while keeping those resources close to the devices. This combination of nearby computing, lower latency, and reduced network traffic is the fundamental benefit of MEC.
4. What Can MEC Be Used For?
MEC is useful for applications that require significant computing resources but also require the computation to be performed close to the user or device.
For example, MEC can be used for:
- Augmented and virtual reality: processing images and other data close to the user.
- Autonomous vehicles: processing sensor and camera data for rapid decision-making.
- Industrial IoT: processing data from machines and sensors within a factory.
- Video analytics: analyzing camera streams without sending all video data to a remote cloud.
- Robotics: performing computationally intensive tasks while keeping the processing close to the robot.
- Smart cities: processing data from cameras, sensors, and other connected devices.
Consider a factory with many cameras monitoring a production line.
Without MEC, the cameras may send their video streams to a remote cloud server:
Cameras
│
│ Video Data
▼
Network
│
▼
Remote Cloud
│
▼
Video Analysis
With MEC, the video can be processed by a MEC server located within or near the factory:
Cameras
│
│ Video Data
▼
MEC Server
│
▼
Video Analysis
The MEC server can analyze the video and send only the required information to the cloud or other systems.
For example, instead of continuously sending an entire video stream, the MEC server may send information such as "defective product detected" or "person detected in restricted area."
MEC is therefore particularly useful when large amounts of data are generated close to users or devices and the application requires the data to be processed quickly.
5. Takeaway
Edge computing is the general concept of bringing powerful computing resources closer to resource-constrained devices, such as mobile phones, robots, cameras, and IoT devices.
MEC (Multi-access Edge Computing) is an implementation of this edge-computing concept within a mobile network. Telecom operators deploy MEC computing resources as part of their network infrastructure, typically close to the mobile access network and users.
In simple terms:
Edge Computing
└── Concept: Bring computing closer to the device
│
▼
MEC
└── Implementation of this concept
within the mobile network by telecom operators
Thus, MEC can be viewed as edge computing implemented as part of the telecommunications network infrastructure, allowing mobile devices to use nearby computing resources instead of relying entirely on remote cloud infrastructure.