Definition of Mobile Edge Computing
Mobile Edge Computing (MEC), now known as multi-access edge computing (multi-access edge computing), its definition and framework come from the European Telecommunications Standards Institute ETSI, its founding members include: HP, Vodafone, Huawei, Nokia, Intel and Viavi etc.
ETSI defines MEC as:
"Provide IT service environment and cloud computing capabilities at the edge of the mobile network", emphasizing that applications, services and content can be deployed locally, in a short distance, and in a distributed manner, thus solving the problem of 5G enhanced mobile broadband, low latency and high reliability to a certain extent And the business requirements of scenarios such as large-scale machine communication terminal connections. With the advent of 5G technology, the application of mobile edge computing will grow exponentially.
The so-called MEC can be understood literally and intuitively. It is a thing that is used in mobile communication systems (Mobile), edge nodes (Edge), and undertakes a large number of computing tasks (Computer).
What is a mobile communication system? We use mobile phones to make calls, send text messages, and surf the Internet, which is called mobile communication. It doesn’t count if we use a network cable to surf the Internet at home, because there is a network cable, and it doesn’t count if we use Wifi to surf the Internet in cafes, because Wifi also needs to be connected to the network cable to access the Internet. What is an edge node? That is, this thing is deployed very close to the user. In technical terms, it is deployed near the base station. What does it mean to undertake a large number of computing tasks? The simplest is the server. Therefore, from the simplest point of view, MEC is a server deployed near the base station of the mobile communication system.
Why should there be MEC? This starts with the mobile communication system. In a traditional mobile communication system, if you want to surf the Internet, the electromagnetic signal emitted by your mobile phone will first be received by the antenna of the base station, and then converted into a digital signal in the base station. Telecom equipment traces back to the core network level by level, and then forwards your request to a specific application server through multiple routes.
In the era when the network speed is still very low and the network performance requirements are still very low, this layer-by-layer forwarding mode is naturally OK, but in a blink of an eye, 5G is coming. 5G has several core rhetoric: ultra-high bandwidth, ultra-high density, and ultra-low latency. What does this mean? In short, what the mobile network has to carry will be dozens or hundreds of times what it used to be, and it can't be delayed.
For example, the principle of unmanned car driving is very simple: the camera on the car shoots video, the server analyzes the road conditions in the video in real time and sends it back to the car in real time. But if it is still forwarded layer by layer like before, then it is estimated that no one will dare to ride in this unmanned car? Why? By the time the steering information is transmitted back to the steering wheel, the car may have crashed.
There are two ways to solve this problem, either to strengthen the capabilities of the network, or to strengthen the capabilities of the front end. The arrival of 5G itself has greatly enhanced the capabilities of the network, but the network has certain uncertainties and is easily disturbed. Therefore, we cannot put all our treasures on certain lines of the network, and the front-end capabilities must also be strengthened.
To strengthen the front end, one is to enhance the processing capability of the camera. Many manufacturers are already doing this. For example, the MPU and VPU promoted by some giants are to add a small processor behind the camera lens. But no matter what, the capability of this kind of small processor is limited after all, and if you want to take big things, you can use hard goods such as Xeon, so another method of front-end enhancement comes: MEC.
The most common MEC deployment is to associate a general-purpose x86 server with the base station to handle urgent and massive processing tasks that need to be processed in the base station. Still taking unmanned vehicles as an example, after the video signal captured by the camera is received by the base station, it will be directly processed in the MEC server, and the driving situation will be adjusted according to the rules, while the remote server will perform tasks such as remote driving and rule update. .
In addition to unmanned vehicles, there are currently a large number of fast computing new services such as AR, VR, high-definition live broadcast, video surveillance, drones, etc., all of which are extremely sensitive to bandwidth and delay. And MEC is to use cheap x86 servers to assist expensive communication components to enhance their capabilities.
Figure: A typical MEC server deployment method
2. The support of 5G technology for the application of digital twins
The so-called 5G is the fifth generation communication technology. In September 2015, the International Telecommunication Union (ITU) officially confirmed the three major application scenarios of 5G, namely eMMB, uRLLC and mMTC.
eMBB is Enhance Mobile Broadband, enhanced mobile broadband. This scenario is an upgraded version of the mobile broadband (mobile Internet access) that people are using now, mainly to serve the needs of consumer Internet. In this scenario, the emphasis is on the bandwidth (rate) of the network.
uRLLC is Ultra Reliable&Low Latency Communication, low-latency, high-reliability communication. This is mainly for IoT scenarios. For example, Internet of Vehicles, UAV, Industrial Internet, etc. In such scenarios, there is a high demand for network latency.
mMTC is Massive Machine Type Communication, massive IoT communication. This is also a typical IoT scenario. For example, smart manhole covers, smart street lights, smart water meters, etc., have a large number of terminals per unit area, and the network needs to support the simultaneous access of these terminals, which refers to the mMTC scenario.
Of these three application scenarios, only one is mainly for the Internet of People, and the other two are mainly for the Internet of Things. Therefore, the Internet of Things attributes of 5G are stronger than the Internet of People attributes.