Control-User Plane Separation (CUPS) is a 3GPP-defined architecture that decouples the control plane functions (managing sessions, mobility, and policy) from the user plane functions (forwarding data packets) in the packet core, allowing them to be scaled, placed, and upgraded independently.
Glossary
Control-User Plane Separation (CUPS)

What is Control-User Plane Separation (CUPS)?
A foundational 5G architecture decoupling session management from data forwarding for independent scaling and placement.
By centralizing control logic while distributing user plane nodes closer to the network edge, CUPS enables energy-efficient network slicing by allowing operators to scale down or sleep idle user plane resources without disrupting session state, directly reducing power consumption in the data plane.
Key Architectural Features of CUPS
Control-User Plane Separation (CUPS) fundamentally restructures the 5G core by splitting the control plane from the user plane, enabling independent scaling, placement, and lifecycle management for optimized energy efficiency and performance.
Independent Scalability
CUPS allows the Session Management Function (SMF) in the control plane and the User Plane Function (UPF) in the user plane to scale independently based on demand.
- Control Plane: Scales with signaling load (session creation, mobility events)
- User Plane: Scales with data throughput (packet forwarding volume)
This prevents over-provisioning of one plane to accommodate the other, directly reducing idle power consumption during low-traffic periods.
Geographic Distribution
The UPF can be deployed at the network edge, close to the user, while the SMF remains in a centralized data center. This minimizes backhaul latency and reduces the energy required to transport user data across long distances.
- Centralized control plane for global policy management
- Distributed user plane for local breakout and edge computing
- Enables Edge Slice architectures for latency-sensitive applications
Packet Processing Optimization
Decoupling enables the UPF to be implemented on specialized, energy-efficient hardware optimized solely for high-speed packet forwarding.
- Accelerator Offloading: UPF functions like GTP-U encapsulation can run on FPGAs or SmartNICs
- Dynamic Voltage and Frequency Scaling (DVFS): Applied independently to UPF workloads based on traffic load
- Eliminates the overhead of running control plane logic on packet processing nodes
Sleep Mode Coordination
CUPS enables precise Sleep Mode Coordination by allowing the control plane to orchestrate power states across distributed user plane instances.
- The SMF can signal idle UPFs to enter deep sleep states
- Wake-Up Signal (WUS) mechanisms can reactivate specific UPFs only when sessions require them
- Enables Cell Discontinuous Transmission (Cell DTX) alignment with core network power states for end-to-end energy savings
Slice-Aware Energy Management
CUPS is foundational for Energy-Efficient Network Slicing. Each network slice instance can have its own dedicated or shared UPF instances, managed independently.
- Slice-Level Energy Models can quantify per-slice power consumption
- Energy-Aware Slice Selection can steer users to slices with the lowest carbon footprint
- Enables Slice Remapping to consolidate traffic onto fewer UPFs during low load, allowing others to power down
Enhanced Resilience and Fault Isolation
A failure in a UPF instance does not necessarily disrupt control plane functions, and vice versa. This isolation improves overall network resilience.
- Control plane can rapidly reassign sessions to healthy UPFs
- Slice Isolation is strengthened by separating control and user plane fault domains
- Enables hitless software upgrades of the SMF without impacting active user data flows
Frequently Asked Questions
Clear, technically precise answers to the most common questions about Control-User Plane Separation in 5G core networks.
Control-User Plane Separation (CUPS) is a 3GPP-defined architectural transformation that decouples the control plane functions (CPFs) from the user plane functions (UPFs) within the packet core, allowing them to be scaled, placed, and managed independently. In a CUPS architecture, the control plane—responsible for session management, mobility, authentication, and policy enforcement—runs on centralized servers, while the user plane—which handles high-speed packet forwarding, classification, and QoS enforcement—can be distributed to the network edge. This separation is achieved by splitting the traditional monolithic gateway (SGW, PGW) into a control plane gateway (SGW-C, PGW-C) and a user plane gateway (SGW-U, PGW-U), connected via the standardized Sx interface using the Packet Forwarding Control Protocol (PFCP). The decoupling means a single control plane instance can manage multiple geographically dispersed user plane nodes, enabling ultra-low latency services by placing UPFs close to end-users while maintaining centralized session control.
Enabling Efficiency, Speed & Accuracy
Intelligent Analysis, Decision & Execution
We build AI systems for teams that need search across company data, workflow automation across tools, or AI features inside products and internal software.
Talk to Us
Search across company data
Give teams answers from docs, tickets, runbooks, and product data with sources and permissions.
Useful when people spend too long searching or get different answers from different systems.

Automate internal workflows
Use AI to route work, draft outputs, trigger actions, and keep approvals and logs in place.
Useful when repetitive work moves across multiple tools and teams.

Add AI to products and internal tools
Build assistants, guided actions, or decision support into the software your team or customers already use.
Useful when AI needs to be part of the product, not a separate tool.
Related Terms
Core architectural components and operational strategies enabled by decoupling the control and user planes in 5G networks.
Session Management Function (SMF)
The control plane function responsible for session establishment, modification, and release. It manages UE IP address allocation, selects and controls the UPF, and enforces policy rules. In a CUPS architecture, the SMF can scale independently from data plane throughput, allowing operators to handle massive signaling loads without over-provisioning packet processing resources.
User Plane Function (UPF)
The data plane workhorse that performs packet routing, forwarding, inspection, and QoS enforcement. Key capabilities include:
- Packet detection rules for classifying traffic into flows
- Usage reporting for charging and policy control
- Buffering downlink packets during UE idle mode CUPS allows UPFs to be placed at the network edge for ultra-low latency while the SMF remains centralized.
Sx Interface (PFCP)
The Packet Forwarding Control Protocol (PFCP) operating over the Sx interface between SMF and UPF. It establishes and maintains the packet detection rules (PDRs) and forwarding action rules (FARs) that govern how the UPF handles traffic. This standardized interface, defined in 3GPP TS 29.244, is what makes multi-vendor CUPS deployments possible, decoupling control logic from forwarding hardware.
Edge UPF Placement
A direct benefit of CUPS where the UPF is deployed at Multi-access Edge Computing (MEC) sites close to the radio network. This minimizes latency for applications like:
- Autonomous vehicle V2X communication
- Industrial AR/VR overlays
- Real-time video analytics The SMF remains in a central data center, maintaining session continuity even as the UE moves between edge UPFs.
Independent Scaling
CUPS enables operators to scale the control plane and user plane independently based on distinct demand patterns. A surge in signaling events—such as massive IoT device registrations—requires more SMF instances without affecting UPF capacity. Conversely, a spike in data throughput from video streaming scales UPFs without wasting control plane resources. This granular elasticity directly reduces infrastructure power consumption.
N4 Interface
The 5G Core equivalent of the Sx interface, connecting the SMF to the UPF. It uses PFCP with extensions for 5G features including:
- QoS flow binding for granular per-flow treatment
- Ethernet PDU session support for industrial protocols
- Redundant transmission for URLLC reliability The N4 interface is the critical control channel that programs the distributed UPF fabric from a logically centralized SMF.

About the author
Prasad Kumkar
CEO & MD, Inference Systems
Prasad Kumkar is the CEO & MD of Inference Systems and writes about AI systems architecture, LLM infrastructure, model serving, evaluation, and production deployment. Over 5+ years, he has worked across computer vision models, L5 autonomous vehicle systems, and LLM research, with a focus on taking complex AI ideas into real-world engineering systems.
His work and writing cover AI systems, large language models, AI agents, multimodal systems, autonomous systems, inference optimization, RAG, evaluation, and production AI engineering.
Partnered with leading AI, data, and software stack.
How We Work
Custom AI workflows for your Business
One-fit-all AI don't work for modern businesses. At Inferensys, we aim to understand your business & custom requirements; which we use to define most efficient agentic workflows, the data, and the tools for your business.
01
Review the use case
We understand the task, the users, and where AI can actually help.
Read more02
Pick the right approach
We define what needs search, automation, or product integration.
Read more03
Build the first useful version
We implement the part that proves the value first.
Read more04
Improve from there
We add the checks and visibility needed to keep it useful.
Read moreThe first call is a practical review of your use case and the right next step.
Talk to Us