Beyond 5G(B5G) and 6 G networks must support diverse applications with widely varying Quality of Service (QoS) requirements, ranging from ultra-low latency to extremely high throughput. Effectively meeting these demands calls for efficient resource management across applications. However, static resource allocation often struggles to adapt to rapid traffic fluctuations and volatile radio channel conditions, resulting in inefficiencies and performance degradation. This requires adaptive, application-aware control that can respond in real time at fine granularity. In this work, we propose a runtime resource control mechanism that enables application-aware resource allocation based on specific QoS requirements. We focus on dynamic Physical Resource Block (PRB) allocation and evaluate its impact on application-level throughput and latency through experimental validation on a B5G testbed, built using open-source RAN (srsRAN) and core network (Open5GS) implementation. Our results show that dynamic PRB allocation has a significant impact on overall performance. We also evaluate the control loop latency to understand the responsiveness and effectiveness of real-time adaptive resource allocation. This study contributes to the open-source community by advancing autonomous resource management, with the implementation released as open-source to support reproducibility.

Remote Direct Memory Access (RDMA) is a key technology in modern data centers, enabling low-latency and high-throughput communication. However, evaluating RDMA performance and validating network designs often requires costly hardware setups or simulation tools with limited performance and realism. In this work, we present P4DMA, a system that leverages programmable switch ASICs to generate realistic high-performance RDMA traffic. By implementing RDMA traffic patterns using the P4 language, P4DMA enables researchers and practitioners to generate RDMA workloads at line rate, without relying on traditional RDMA NICs (RNICS). With Tofino's traffic generation capacity of up to Tbps, P4DMA offers a novel approach to stress and evaluate RDMA-capable infrastructures, and accelerate the prototyping of new RDMA-based applications.

This paper presents an in-network machine learning (ML) approach for classifying Augmented Reality (AR) and Cloud Gaming (CG) traffic using programmable hardware. Random Forest (RF) models are deployed in a P411P4: Programming Protocol-independent Packet Processors data plane capable of processing Real-time Transport Protocol (RTP) traffic features like Frame Size (FS) and Inter-Frame Interval (IFI) for efficient classification. The classifier marks AR and CG traffic with Explicit Congestion Notification (ECN) codepoints to integrate with the Low Latency, Low Loss, Scalable Throughput (L4S) features of the programmable switch. The RF model prioritizes AR/CG traffic using Differentiated Services Code-Point (DSCP) assignments and modular ECN marking. The classification performance is evaluated using accuracy, precision, recall, and F1-score, while time overhead is assessed based on nodal processing time incurred during deployment by replaying AR/CG traffic. The P4 implementations for P4Pi22https://eng.ox.ac.uk/computing/projects/programmable-hardware/p4pi.(V1Model) and Tofino Native Architecture (TNA) are all publicly available.

Augmented Reality (AR) real-time interaction between users and digital overlays in the real world demands low latency to ensure seamless experiences. To address computational and battery constraints, AR devices often offload processing-intensive tasks to edge servers, enhancing performance and user experience. With the increasing adoption and complexity of AR applications, especially in remote rendering, accurately classifying AR network traffic becomes essential for effective resource allocation. This paper explores two methods based on Decision Tree (DT) and Random Forest (RF) to classify network traffic among AR, Cloud Gaming (CG), and other categories. We rigorously analyze specific features to precisely identify AR and CG traffic. Our models demonstrate robust performance, achieving accuracy rates ranging from 88.40% to 94.87% against pre-existing datasets. Moreover, we contribute with a novel dataset encompassing AR and CG traffic, curated specifically for this study and made publicly available to facilitate reproducible research in AR network traffic classification.

This paper focuses on hybrid pipeline designs for User Plane Function and next-generation NodeB leveraging target-specific features and an insightful discussion of P4 and target challenges and limitations. The entire or disaggregated UPF runs on P4 targets and allocates packet processing data paths in P4 hardware or DPDK/x86 software based on flow characteristics (e.g., heavy hitters) and QoS requirements (e.g., low-latency slices). For the hybrid gNodeB, most packet processing is executed in commodity Tofino hardware, while unsupported functions such as Automatic Repeat Request and cryptography are performed in DPDK/x86. We show that our hybrid UPF improves the scalability by 18× and reduces latency up to 50%. The results also suggest that careful traffic allocation to pipeline targets is required to optimize each target's strength and avoid processing delays. Finally, we demonstrate a QoS-oriented application of the hybrid UPF and present gNodeB buffer service benchmarks.

The research community has recently proposed several solutions based on modern programmable switches to detect entirely in the data plane the flows exceeding pre-determined threshold in a time window, i.e., Heavy Hitters (HH). This is commonly achieved by dividing the network stream into fixed time slots and identifying each separately without considering the traffic trends from previous intervals. In this work, we show that using specified time windows can lead to high inaccuracies. We make a case for rethinking how switches analyze the incoming packets and propose to leverage per-flow Inter Packet Gap (IPG) analytics instead of using flow counters for HH detection. We propose an algorithm and present a P4 pipeline design using this new metric in mind. We implement our solution on P4 hardware and experimentally evaluate it against real traffic traces. We show that our results are more accurate than related work by up to 20% while reducing the control channel overhead by up to two orders of magnitude. Finally, we showcase a QoS-oriented application of the proposed dataplane-only IPG-based HH detection in a mobile network scenario.