Edge Computing

Description

Edge Computing, as defined in 3GPP TS 23.501 and further elaborated in TS 23.558 and TS 28.538, is a fundamental concept in 5G and beyond that brings computation and storage resources closer to the user equipment (UE). The definition states: 'A concept, as described in 3GPP TS 23.501, that enables operator and 3rd party services to be hosted close to the UE's access point of attachment, to achieve an efficient service delivery through the reduced end-to-end latency and load on the transport network.' This approach is critical for applications requiring ultra-low latency, such as augmented reality, autonomous vehicles, and industrial IoT.

The architecture for Edge Computing in 3GPP is built around the Edge Application Server (EAS) and the Edge Enabler Server (EES), as specified in TS 23.558. The EAS hosts the application logic and runs at the edge, while the EES provides supporting functions such as service provisioning, registration, and discovery. The Edge Enabler Client (EEC) resides in the UE and communicates with the EES to discover and connect to available EASs. The 5G core network, particularly the User Plane Function (UPF), plays a key role by enabling local traffic offloading through mechanisms like Uplink Classifier (ULCL) and IPv6 Multi-homing, allowing traffic to be routed to edge applications without traversing the central core. The Network Exposure Function (NEF) and Policy Control Function (PCF) provide APIs and policies for edge computing service provisioning.

How it works: When a UE requests a service that can be served by an edge application, the network determines the appropriate EAS based on factors like UE location, service requirements, and EAS availability. The EES assists in EAS discovery and provides the UE with the necessary information to connect. The UPF is configured to route traffic to the local data network where the EAS resides, often using a local breakout. This reduces latency by avoiding long backhaul to centralized data centers. The 3GPP management system (TS 28.538) specifies how edge computing resources are managed, including lifecycle management of EASs, performance assurance, and fault management. The EES and EAS can be owned by the operator or third parties, and 3GPP defines interfaces for interworking.

Key components include: the EAS (hosts the application), EES (provides edge enabler functions), EEC (client in UE), UPF (user plane anchor for local offloading), NEF (exposes edge services to third parties), and the management system (for orchestration). The architecture supports both operator-managed and third-party edge clouds, enabling a multi-access edge computing (MEC) ecosystem. 3GPP also defines Edge Computing in the context of network slicing, where a slice can be dedicated to edge services.

Edge Computing is pivotal for 5G services because it enables new revenue streams for operators and enhances user experience. It supports use cases like cloud gaming, smart manufacturing, and connected vehicles. By reducing latency and backhaul load, it improves network efficiency. As networks move towards 6G, edge computing will become even more integrated with AI/ML and network automation, enabling real-time analytics and decision-making at the edge. The 3GPP specifications ensure standardized interfaces and management, facilitating interoperability and widespread adoption.

Purpose & Motivation

Edge Computing was created to overcome the limitations of centralized cloud computing in meeting the stringent latency and bandwidth requirements of emerging 5G applications. Traditional cloud architectures host services in distant data centers, resulting in high end-to-end latency and significant load on the transport network. For applications like autonomous driving, industrial automation, and AR/VR, such latency is unacceptable. Edge Computing solves this by moving compute and storage resources closer to the UE, at the edge of the network.

Historically, the concept evolved from Multi-access Edge Computing (MEC) standardized by ETSI, which focused on placing applications at the network edge. 3GPP adopted and extended this concept in Release 15 and 16, defining the Edge Application Server and Edge Enabler Server in TS 23.558 (Release 17). The need for standardized edge computing in 3GPP arose from the desire to integrate edge services seamlessly with 5G core network functions, ensuring optimal routing, policy control, and service continuity.

The motivation includes: (1) reducing latency for time-critical applications; (2) offloading traffic from the core network to reduce congestion; (3) enabling new services and business models for operators and third parties; (4) improving user experience through localized content delivery; and (5) supporting network slicing with dedicated edge resources. By standardizing edge computing, 3GPP ensures that different vendors and operators can interoperate, accelerating deployment and innovation.