THE LINUX FOUNDATION PROJECTS

8th European P4 Workshop (EuroP4’26)

The 8th edition of EuroP4 will be on December 7, 2026, in co-location with CoNEXT’26 (December 7-10), in Utrecht, The Netherlands.

EuroP4’26 brings together networking researchers and practitioners to discuss cutting-edge, P4-enabled research and P4-based technology. The workshop provides a venue for presenting and discussing research and projects related to P4, as well as for more broadly discussing the needs of this research community. The workshop aims to forge new connections between researchers who already work with P4, introduce more networking researchers to the P4 community, and seed future top-tier publications, innovation, and contributions to this community.

You can register for our Workshop on the CoNEXT website. “Workshop-Only” passes are available.

Register Here
EuroP4 2026 Logo

Workshop Program

Each time slots include time for Q&A.

Welcome

9:00–9:10

Welcome and Chairs’ Introduction

9:10–10:10

Keynote | In-network ML: Inference at the Speed of Data

Speaker: Noa Zilberman (Oxford University)

10:10–10:25

Poster discussion and demo: XNA: Programming the X2 with P4

Authors: Fabian Ruffy, Xsight Labs; Vladimir Gurevich, Xsight Labs, Manolis Tsamis, Compiler Tool Chain Consulting; and Pietro Ghiglio, Compiler Tool Chain Consulting

10:25–11:00

Coffee Break

Session 1: Traffic measurement, control, and network infrastructure

11:00–11:15

WFCM: Accelerating Bayesian Flow Size Distribution Estimation with Flow

Authors: Joep Boumans, Eindhoven University of Technology and Habib Mostafaei, Eindhoven University of Technology

11:15–11:30

Stateless Egress Protection Using MPLS Network Actions: Concept and P4-Based Hardware Implementation

Authors: Fabian Ihle, Etienne Zink, and Michael Menth, all from University of Tuebingen

11:30–11:45

SCION-IP Translation at Scale with Vitis Networking P4 and OpenNIC Shell

Authors: Lars-Christian Schulz, Aneesh Pindalavalappil Sathyan, Robin Wehner, and David Hausheer, all from Otto-von-Guericke-University

11:45–12:00

Poster discussion and demo: Expanding TFTG: Multi-flow Generation and Credit-Based Shaping for Deterministic Networks

Authors: Sérgio Rossi Silva, University of Campinas; Francisco G. Vogt, University of Campinas’ Zhiheng Yang, University of Amsterdam; Chrysa Papagianni, University of Amsterdam; Flavio G. Rocha, Universidade Federal de Goias; and Christian E. Rothenberg, University of Campinas

12:00–14:00

Lunch Break

14:00–15:00

Keynote | From Programmable Networks to Programmable Testbeds for High-Fidelity Experimentation

Speaker: Christian Esteve Rothenberg (University of Campinas)

Session 2: P4 Verification and Compilation

15:00–15:15

P4Check: Verification and Debugging for Stateful P4 Programs

Authors: Tommaso Pacciani, L. Thomas van Binsbergen, and Chrysa Papagianni, all from University of Amsterdam

15:15–15:30

Compiling P4 to Hot-Swappable NF Graph Stages

Authors: Károly Kecskeméti, ELTE Eötvös Loránd University; Sándor Laki, ELTE Eötvös Loránd University; and Géza Szabó, Ericsson Hungary Ltd.

15:30–16:00

Coffee Break

Session 3: In-Network Machine Learning

16:00–16:15

PlanPi: In-Network Machine Learning via p4c-dpdk

Authors: Yuzhong Luo, Peng Qian, and Noa Zilberman, all from University of Oxford

16:15–16:30

Adaptive In-Network Random Forests: A Deployable Lifecycle for Programmable Switches

Authors: Juled Zaganjori, Technical University of Munich; Elmer Dema, University of Augsburg; Lukas Froschauer, Deggendorf Institute of Technology; and Andreas Kassler, Deggendorf Institute of Technology; Karlstad University

Panel and Closing

16:30–16:55

Panel discussion

16:55–17:00

Closing Remarks

Keynote Sessions

In-network ML: Inference at the Speed of Data

As network speeds continue to increase and applications demand ever-faster decisions, moving data to remote compute resources for machine learning (ML) inference introduces latency and scalability challenges. Programmable data planes offer an alternative: executing ML inference directly on packets as they traverse the network at line rate.

However, mapping ML models onto programmable switches is far from straightforward. Limited memory, restricted arithmetic operations, and stringent pipeline constraints pose significant challenges to supporting expressive models while preserving throughput and accuracy.

In this talk, I will trace our research in in-network ML, from early foundational work to practical and accessible open-source solutions. I will introduce tools and frameworks that provide automated model translation, resource-aware deployment, and distributed inference across multiple network devices. These resources enable applications in domains such as finance, cybersecurity, manufacturing, and healthcare that benefit from real-time, in-network decision-making.

Finally, I will discuss emerging research directions and the opportunities and challenges of communicating in-network ML solutions to the wider practitioner community.

Noa Zilberman

Noa Zilberman

Oxford University

From Programmable Networks to Programmable Testbeds for High-Fidelity Experimentation

Programmable network devices were originally conceived as platforms for implementing new packet-processing functions. But what if we programmed them not only to build the network under study, but also to build the testbed around it?

This keynote explores “Software-Defined Inception”: the use of programmable switches and SmartNICs as building blocks for realistic, reproducible network experimentation. It presents an evolving open-source toolbox that combines hardware-accelerated network emulation, traffic generation, and protocol testing. Tools such as the P4 Programmable Patch Panel (P7), PIPO-TG, and P4R demonstrate how programmable hardware can reproduce configurable topologies and link conditions, replay packet traces, and generate challenging traffic patterns—including stateful flows—at line rate.

Drawing on experience with these tools in research and education, the talk discusses both the opportunities and practical challenges of turning data-plane devices into experimental infrastructure. These include preserving timing fidelity, sharing limited hardware resources, composing experimental functions with the system under test, and making experiments portable and reproducible across platforms.

Finally, the talk offers a first look at SPINE, ongoing work toward a practical network tester spanning programmable switches and SmartNICs. SPINE builds on the lessons of the existing toolbox to explore how limited physical infrastructure can represent richer experimental environments while retaining hardware-level fidelity.

The broader argument is that the next frontier for programmable data planes extends beyond implementing individual network functions. By programming the experimental environment itself, we can create compact, flexible, and high-performance testbeds in which future network protocols and systems can be built, stressed, observed, and validated under realistic conditions.

Christian Esteve Rothenberg

Christian Esteve Rothenberg

University of Campinas

Important Dates

  • Submission deadline: Friday, August 7, 2026 at 11:59 pm CET
    Monday, August 17, 2026 at 11:59 pm CET (Hard deadline)
  • Posters and demos deadline: Tuesday, September 1, 2026
    Sunday, September 6 at 11:59 pm CET (Hard deadline)
  • Acceptance notification: Monday, September 14, 2026
  • Posters and demos notification: Monday, September 14, 2026
  • Camera-ready: Monday, September 28, 2026

Contact Us

If you have any questions, please email the TPC co-chairs:

  • h.mostafaei_at_tue.nl
  • vaddank_at_purdue.edu

Chairs and Committee Members

Here are this year’s chairs and committee members.

General Chairs

  • Chrysa Papagianni, University of Amsterdam
  • Davide Scano, Independent

Program Committee Chairs

  • Habib Mostafaei, Eindhoven University of Technology
  • Vamsi Addanki, Purdue University

Steering Committee

  • Muhammad Shahbaz, University of Michigan
  • Fernando Ramos, University of Lisbon

Publicity Chair

  • Leandro Almeida, Federal Institute of Education, Science and Technology of Paraíba

Technical Program Committee

  • Alan Zaoxing Liu (University of Maryland)
  • Christian E. Rothenberg (University of Campinas)
  • Fábio Luciano Verdi (Federal University of São Carlos)
  • Harsh Gondaliya (Meta)
  • Jinhua Guo (University of Michigan-Dearborn)
  • Noa Zilberman (University of Oxford)
  • Mariano Scazzariello (RISE)
  • Masoud Hemmatpour (COM4AI)
  • Mingyuan Zang (China Telecom, Technical University of Denmark)
  • Nik Sultana (Illinois Institute of Technology)
  • Salvatore Signorello (NOVA University Lisbon)
  • Sandor Laki (ELTE Eötvös Loránd University)
  • Tommaso Caiazzi (Roma Tre University)
  • Vladimir Gurevich (XSight)
  • Xiaoqi Chen (Purdue University)