On the Effect of TSN Forwarding Mechanisms on Best-Effort Traffic

TL;DR

This work addresses how Time-Sensitive Networking forwarding mechanisms affect best-effort traffic in mixed real-time networks. It surveys SP, TAS, CBS, ATS, and ETS, analyzing their impact on BE delays and queueing using simulations in synthetic and automotive topologies. The results show that ATS, CBS, and ETS significantly reduce BE delay and backlog compared with SP and TAS, with up to ~20× improvement under certain configurations, and ETS often performing best at higher high-priority utilization $U_h$. These findings offer practical guidance for designing TSN networks that balance real-time guarantees with BE performance across diverse traffic scenarios.

Abstract

Time-Sensitive Networking (TSN) enables the transmission of multiple traffic types within a single network. While the performance of high-priority traffic has been extensively studied in recent years, the performance of low-priority traffic varies significantly between different TSN forwarding algorithms. This paper provides an overview of existing TSN forwarding algorithms and discusses their impact on best-effort traffic. The effects are quantified through simulations of synthetic and realistic networks. The considered forwarding mechanisms are Strict Priority (SP), Asynchronous Traffic Shaper (ATS), Credit-Based Shaper (CBS), Enhanced Transmission Selection (ETS), and Time-Aware Shaper (TAS). The findings indicate that ATS, CBS, and ETS can significantly reduce queuing delays and queue lengths for best-effort traffic when compared to SP and TAS. This effect is enhanced when the reserved bandwidth for high priority queues - using CBS, ATS, or ETS - is reduced to the lowest possible value, within the reserved rate and latency requirements. Specifically, the simulations demonstrate that the choice of forwarding algorithm can improve the performance of low-priority traffic by up to twenty times compared to the least effective algorithm. This study not only provides a comprehensive understanding of the various TSN forwarding algorithms but can also serve as guidance at networks' design time to improve the performance for all types of traffic.

Figures (11)

• Figure 1: TSN forwarding logic per egress port
• Figure 2: Credit evolution for CBS forwarding
• Figure 3: Different forwarding algorithms compared
• Figure 4: Different effects that reduce the effective bandwidth when using TAS
• Figure 5: Evaluations on different configuration in one-hop network