This project investigated various performance aspects of the control plane in modern, software-defined networks. The focus was on real-time networks that can provide deterministic latency bounds for high-priority traffic. Central research questions included flexibility and scalability while maintaining guarantees, comparisons between centralized and decentralized architectures, as well as guidelines for evaluating and selecting solution approaches. At the start of the project, there was no established methodology for assessing the scalability of such systems. Therefore, during the course of the project, the Scalability Index was developed and used as a general methodology for comparing two systems. In addition, meaningful metrics for evaluating the performance of real-time networks were defined. Reliable measurements are a key aspect of real-time networks, so a measurement methodology with sub-microsecond accuracy was developed, which can be implemented using standard consumer components. These methods were used to analyze the performance of real-time networks from various perspectives. Centralized and decentralized control approaches were designed to flexibly handle requests for real-time transmissions. The response times of centralized and decentralized approaches were also compared using simulations and analytical models. The influence of topological properties on network efficiency was traced back to general graph metrics. Finally, the approaches were evaluated in a broader context, including in conjunction with other real-time concepts and the 5G ecosystem. Overall, the scientific toolkit for evaluating real-time networks was significantly expanded, and guidelines for their performance assessment were developed.
Building similarity graph...
Analyzing shared references across papers
Loading...
Alexej Grigorjew
David Raunecker
Viktoria Vomhoff
Building similarity graph...
Analyzing shared references across papers
Loading...
Grigorjew et al. (Thu,) studied this question.
www.synapsesocial.com/papers/69abc1535af8044f7a4e9ee2 — DOI: https://doi.org/10.25972/opus-44032