Serving Gateway Control plane function

Description

The Serving Gateway Control plane function (SGW-C) is a critical network function introduced as part of the Control and User Plane Separation (CUPS) architecture for the Evolved Packet Core (EPC) and its evolution into 5G. It represents the disaggregation of the monolithic Serving Gateway (SGW) into separate control and user plane entities. The SGW-C is responsible for all the control logic associated with managing user sessions. This includes processing signaling from the Mobility Management Entity (MME) or Access and Mobility Management Function (AMF), such as session establishment, modification, and termination requests. It manages the mobility anchoring point for inter-eNodeB handovers within LTE and acts as the local mobility anchor for 3GPP access when the user equipment (UE) moves between eNodeBs. The SGW-C also interfaces with the Policy and Charging Rules Function (PCRF) or Policy Control Function (PCF) to enforce QoS policies and charging rules for each data session.

Architecturally, the SGW-C communicates with its corresponding user plane function, the SGW-U, using the Packet Forwarding Control Protocol (PFCP) as defined in 3GPP TS 29.244. This protocol allows the SGW-C to instruct the SGW-U on how to handle user plane packets. For example, the SGW-C sends PFCP Session Establishment/Modification/Deletion requests to the SGW-U to create, update, or remove packet detection rules (PDRs), forwarding action rules (FARs), QoS enforcement rules (QERs), and usage reporting rules (URR). This separation allows the SGW-C to be deployed centrally for efficient control, while SGW-U instances can be distributed to the network edge to reduce latency for user traffic.

In a 5G context, the SGW-C is part of the interworking function between the 5G Core (5GC) and the EPC, particularly in non-standalone (NSA) deployment scenarios. It works in conjunction with the User Plane Function (UPF) and the Session Management Function (SMF). The SGW-C's role is to maintain continuity for 4G sessions as networks transition to 5G, ensuring seamless mobility and service. Its operation is detailed across multiple specifications, including architecture (23.214), management (28.708, 32.867), protocol (29.244), and security (33.127).

Purpose & Motivation

The SGW-C was created to address the limitations of the traditional, integrated Serving Gateway in 4G networks. The monolithic SGW combined both control and user plane functions in a single network element, which led to scaling inefficiencies. Scaling for increased control signaling (e.g., during mass device attach events) required scaling the entire box, including expensive user plane packet processing resources. Conversely, scaling for user plane throughput (e.g., during a video streaming surge) necessitated scaling the control plane capacity as well, leading to cost-ineffective and rigid deployments.

The introduction of the SGW-C, as part of the CUPS framework standardized in 3GPP Release 14, was motivated by the need for greater network flexibility, scalability, and innovation. By separating the control plane, operators can centralize and pool SGW-C resources in large data centers for efficient management and signaling processing. This separation enables independent scaling of control and user plane resources based on actual network demands. It also facilitates the deployment of user plane functions (SGW-U) at distributed locations closer to the radio access network, which is a key requirement for low-latency applications and edge computing. This architectural shift was a foundational step towards the fully cloud-native, service-based architecture of the 5G Core network.