5G Mobile Edge Computing (MEC), also known as Multi-access edge computing, is a network architecture that combines elements of both IT technology and cloud computing to provide a software network framework at the edges of a 5G cellular network.
Implementing MEC at the edge nodes of a 5G network may provide efficient network utilization of computing and communications resources. This will lead to off-load network processes and reduce overall network congestion.
5G MEC may increase user experience by running applications closer to the 5G cell user. Moving the content closer to the user may result in lower end-to-end latency. MEC may, therefore, help Mobile Operators to develop new 5G applications and new 5G services in a more flexible and scalable manner.
5G MEC concept
As technology evolves, 5G telecom vendors are developing devices that require more data. In a new 5G network architecture, the number of connected devices can increase exponentially. Each network node will have different requirements depending on the purpose they serve and on their geographical location. Also, as a general rule, network elements that need to run real-time applications have stringent delay constraints.
5G network devices are part of a broader ecosystem that requires computing power and storage to process and store data. It is estimated that tens of billions of such Edge devices will be deployed shortly. The processing speeds will also multiply.
Cloud computing is the first solution for finding these resources. However, when the goal is to develop a platform to support a wide range of IoT applications, the level of requirements makes even the Cloud challenging to handle. Therefore, by accumulating the enormous amount of inactive computing power and storage distributed across the network, Mobile Edge Computing can provide sufficient capacity to perform computationally sensitive and critical delay tasks on mobile devices.
Traffic delay remains the main challenge for 5G URLLC services. Therefore, 5G MEC is widely accepted to be a key technology for delivering various applications for the next generation of the Internet, such as Tactile Internet and the Internet of Things (IoT).
5G MEC technology aspect
Currently, researchers both from academia and the industry have been turning to MEC technology, seeking ways to take advantage of both mobile computing and wireless communications.
MEC and the transfer of resources to the edge was not designed to compete with the Cloud, but to complement and extend it. Smaller servers can now be placed at the edge of the network. This way, a platform provider may deploy endpoints, at the Cloud or the edge, as long as application requirements and restrictions are met. All of these technologies need to successfully interoperate to provide smart communication, orchestration, and resource allocation designed to meet IoT requirements.
MEC offers cloud computing capabilities within the RAN. Instead of allowing direct mobile traffic between the backbone network and end-users, MEC connects the user directly to the nearest service cloud at the edge of the network. Developing MEC at the base station improves computation processing and avoids congestion and system downtime.
5G MEC advantages
On the 5G network side, reducing end-to-end delay and the load on transport networks are the two principal objectives. The amount of data that needs to be transferred is significantly reduced with Edge application services. MEC also improves the quality of service (QoS) due to the reduced distance of the data flow. Consequently, transmission costs and service latency are also reduced.
Data bottlenecks and potential failure points are restricted because the core of the computing environment is not actively involved in the data flow process. Encrypted data is moved to the central network, thereby improving security. Still, the data passes through firewalls and is checked for viruses even at the edge of the system.
5G MEC: Moving from Cloud to the Edge
Over the last years, the popularity of mobile devices and the exponential growth of Internet traffic have driven various innovations in wireless communications and networking. In particular, 5G small cell networks, 5G active antennas, and millimeter-wave communications promise to provide users with gigabit speeds.
The high data rates and extremely reliable air interface allow mobile computing services to run on remote data centers. In the research field, this is known as Mobile Cloud Computing (MCC). However, one limitation of MCC is the long-distance transmission from end-users to remote data centers, which may lead to excessive waiting times in mobile applications. Currently, MEC architecture is being designed to integrate the concept of Cloud Computing into 5G cellular networks.
MEC is implemented based on virtualized platforms taking advantage of recent developments in Network Functionalization (NFV), Information-centric networks, ICNs and Software-defined Networking (SDN). In particular, NFV may activate a simple edge device to provide computing services to multiple mobile devices by creating virtual machines (VMs) and perform different tasks or different network functions simultaneously.
On the other hand, ICN provides an alternative example of end-to-end service recognition for the MEC, shifting from host-centric to information-centric and implementing environment-based computations. Finally, SDN enables MEC network administrators to manage services through removable functions, achieving scalable, and dynamic computing. The main focus of MEC research is to develop further these generic network technologies so that they can be implemented in the 5G network edge.
5G MEC standardization
MEC technical standards have been mainly developed by the European Telecommunications Standards Institute (ETSI). Cisco has also proposed a new concept as a general form of MEC where the definition of edge devices is expanded further, ranging from smartphones to set-top boxes. This leads to a new research area called Fog Computing and Networking. However, the fields of Fog Computing and MEC overlap and the terminology is often used interchangeably.
5G networks allow connectivity across a wide range of devices to support new features and functionalities. To achieve the above requirements 5G networks need to transfer functionalities from hardware to software. These software-based functions have been developed thanks to the recently established SDN and NFV techniques. Since 2015, MEC (along with SDN and VFN) has been recognized by 5GPPP as one of the key emerging technologies for 5G networks. Since 2017, 3GPP has also included edge computing support, as one of the critical features in 5G systems. 5G MEC poses as a natural evolution of mobile base stations and the convergence of IT and networking telecommunications to support the new 5G ecosystem.
