Design and Evaluation of Flexible Programmable Hybrid Real-time Networks with Hard and Soft Real-time Guarantees

Funding Agency: DFG
Duration: 3 years, 01.11.2021 - 31.10.2024
Partners: Lehrstuhl für Informatik III, Universität Würzburg
Researchers (TUM/LKN): Philip Diederich (
  Dr. Andreas Blenk (
  Prof. Wolfgang Kellerer (
Researchers (WUE/LS3): Stefan Geißler
  Prof. Tobias Hoßfeld


Scope of Project

The use of software-defined networking (SDN) allows for integrating new mechanisms for the joint optimization of application control and network control. Phase 1 of the DFG SDN-APP project investigated and showed the feasibility of application-aware network management with an improved Quality of Experience (QoE) in multi-application scenarios.

Phase 2 of the DFG SDN App project extends Phase 1 and widens the considered scenarios. It integrates time-critical applications into the scope of the project. Phase 2 focuses on hybrid real-time networks with hard and soft real-time guarantees with changing demands like smart manufacturing to control manufacturing devices.

Phase 2 of the DFG SDN-APP project aims at realizing a hybrid real-time network allowing time-critical and best-effort applications to coexist on the same infrastructure. Phase 2 investigates the use of concepts such as Time-Sensitive Networking (TSN) and Network programmability (SDN and P4) for hybrid real-time networks.

The main objectives cover:

  • Hybrid real-time networks: Mechanisms to provide hard and soft real-time requirements on a common physical infrastructure for (industrial) network types by utilizing SDN-enabled application-aware network control architectures
  • Programmability:
    • Analysis of programmable networking hardware with a focus on reconfiguration
    • Identification of missing features
    • Adaptation of the algorithms for a “hardware-aware” solution for providing end-to-end guarantees
  • Flexibility: Quantification of the flexibility, as well as the performance/efficiency increase, depending on the available information
  • Guidelines: Guidelines for the given network type



This work received funding by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) - 316878574.