Chien-Chao Tseng

The 5G Wireless and Wireline Convergence (5G-WWC) standards introduce critical network functions—notably the Access Gateway Function (AGF) and the User Plane Function (UPF)—to enable unified wired and wireless access through a single 5G core. However, deploying and integrating these functions across heterogeneous programmable hardware platforms remains a significant open architectural challenge. This paper presents a configurable integration framework that orchestrates AGF and UPF workloads on heterogeneous programmable data planes, specifically NVIDIA BlueField-2 Data Processing Units (DPUs) and P4-based switches. Unlike traditional, hardware-specific implementations, the framework provides a unified control plane that dynamically manages AGF-only, UPF-only, or Combined AGF/UPF deployments. A hardware abstraction mechanism decouples the control logic from pipeline-specific details, enabling the same control plane to drive different underlying hardware without modification. A Generic Flow Rule interface standardises communication between the control plane and each user-plane backend, while a merged DPU pipeline for Combined AGF/UPF eliminates the redundant GTP-U encapsulation and decapsulation steps inherent in a naively cascaded design. Experiments on NVIDIA BlueField-2 DPUs achieve near-100 Gbps throughput across all three TR-470 scenarios (AGF-only, UPF-only, and Collocated AGF/UPF). The Combined AGF/UPF configuration exhibits lower end-to-end latency than the separated AGF + UPF configuration, confirming both the feasibility and the efficiency of the proposed framework for next-generation high-performance programmable networks.

The 3rd Generation Partnership Project (3GPP) proposed a new non-3GPP access network, the Trusted WLAN Access Network (TWAN), in Release 16 of the 5th Generation Mobile Communication Network (5G). TWAN enables devices without 5G capability to access 5G networks via WLAN, allowing operators to serve non-5G devices in the market through Wi-Fi services. This paper divides TWAN into a control plane and a user plane, which are handled by an SDN controller and P4 switches, respectively. This setup, named P4-TWAN, is further integrated with User Plane Function (UPF) to eliminate unnecessary tunneling. Experimental results show that P4-TWAN processed user traffic at line rate while the integration with UPF benefited from the elimination of unneeded tunneling.

Network slicing is an essential technology for 5G mobile networks. It partitions network resource logically into multiple isolated slices, each of which can satisfy a suite of network requirements for one specific service. However, it cannot be fulfilled by the current SDN (Software-Defined Networks), since the conventional SDN data-plane technology, OpenFlow, is not flexible enough to offer fine-grained network resource control or queue/packet scheduling. It leads to many research studies developing corresponding solutions on programmable switches. In this work, we focus on the support of the bandwidth guarantee and management for network slices. Although several studies with the similar goal have been proposed, they do not consider interference among different flow types or use the built-in meter for easy deployment on COTS (Commercial Off-The-Shelf) P4 switches. To this end, we first conduct a case study to examine the interference cases. We then propose a solution, designated as P4-TINS (P4-driven Traffic Isolation for Network Slicing), to resolve the interference by isolating different types of traffic flows in priority queues and set the P4 switch’s bucket size based on the time granularity of its bandwidth management operation. It cannot only ensure the guaranteed bandwidth for each slice but also enable coexistent slices to fairly share residual bandwidth. We have confirmed its effectiveness experimentally based on our prototype over an ONOS (Open Network Operating System) controller and a COTS P4 switch.

An SDN switch newly added to a network needs to establish a control channel with an SDN controller to manage control-plane traffic. The setup of such a channel is termed bootstrapping. In the case of in-band control, bootstrapping involves setting up a control path that may traverse one or more switches between the new switch and the controller, which is achieved through the configuration of layer-two to layer-four parameters on relevant switches. Previous approaches either result in lengthy bootstrapping time due to their level-by-level mechanisms, demand complicated modules, or lack some essential features. This paper proposes an approach to fast automatic bootstrapping that overlaps the bootstrapping processes among a set of switches. Our emulations using virtual switches completed the process of bootstrapping 50 switches arranged in a chain in 2.0 seconds, which is a 98% time reduction compared with a prior study.

Network slicing is considered a key technology in enabling the underlying 5G mobile network infrastructure to meet diverse service requirements. In this article, we demonstrate how transport network slicing accommodates the various network service requirements of Massive IoT (MIoT), Critical IoT (CIoT), and Mobile Broadband (MBB) applications. Given that most of the research conducted previously to measure 5G network slicing is done through simulations, we utilized SimTalk, an IoT application traffic emulator, to emulate large amounts of realistic traffic patterns in order to study the effects of transport network slicing on IoT and MBB applications. Furthermore, we developed several MIoT, CIoT, and MBB applications that operate sustainably on several campuses and directed both real and emulated traffic into a Programming Protocol-Independent Packet Processors (P4)-based 5G testbed. We then examined the performance in terms of throughput, packet loss, and latency. Our study indicates that applications with different traffic characteristics need different corresponding Committed Information Rate (CIR) ratios. The CIR ratio is the CIR setting for a P4 meter in physical switch hardware over the aggregated data rate of applications of the same type. A low CIR ratio adversely affects the application’s performance because P4 switches will dispatch application packets to the low-priority queue if the packet arrival rate exceeds the CIR setting for the same type of applications. In our testbed, both exemplar MBB applications required a CIR ratio of 140% to achieve, respectively, a near 100% throughput percentage with a 0.0035% loss rate and an approximate 100% throughput percentage with a 0.0017% loss rate. However, the exemplar CIoT and MIoT applications required a CIR ratio of 120% and 100%, respectively, to reach a 100% throughput percentage without any packet loss. With the proper CIR settings for the P4 meters, the proposed transport network slicing mechanism can enforce the committed rates and fulfill the latency and reliability requirements for 5G MIoT, CIoT, and MBB applications in both TCP and UDP.

Software-defined network (SDN) allows traffic engineering and flow management by decoupling the control and data planes. SDN has attracted a lot of attention recently because of its centralized control, flow programmability and flexible resource management. However, to support centralized control, flow programmability and flexible resource management, most of the advanced SDN designs require per-flow management, which would easily make the flow table of a software switch overflow and incur undesired processing overhead when a network scales up. To overcome such huge traffic demands, we propose a cluster-based macro-flow management framework to balance the trade-off between customized services and management costs. The key idea of our design is to group similar flows into a cluster and replace per-flow management with per-cluster management. To achieve this goal, we cluster flows with consideration of the similarity of their traffic patterns. To adapt to varying traffic load, we further propose a hierarchical cluster merging scheme to dynamically merge clusters and make the best tradeoff between flow entry saving and routing performance guarantee. Our evaluation demonstrates that the proposed cluster-based flow management framework can significantly reduce the flow table usage, while producing minimum negative impact on routing performance.