Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploiting Dependency-Aware Priority Adjustment for Mixed-Criticality TSN Flow Scheduling

Miao Guo, Yifei Sun, Chaojie Gu, Shibo He, Zhiguo Shi

TL;DR

This paper tackles the queuing delays that arise in mixed-criticality TSN when using static priorities. It introduces a dependency-aware adaptive priority adjustment method that operates across TT, AVB-A, and AVB-B flows, accounting for different shaping mechanisms and link overlaps. Two algorithmic regimes are proposed: Parallel-FG, where routes are disjoint, and Cross-FG, where FG conflicts are considered via TTConfliction and FG-conflict graphs with metrics TTAdjust and AVBAdjust; these yield schedules that maximize the number of flows meeting their deadlines under RTA-based analysis. Experimental results on realistic SAE, AFDX, and Ladder topologies show schedulability improvements up to about 20–29% over state-of-the-art baselines, along with higher link utilization and substantially lower computation times, demonstrating practical gains for real-time TSN systems.

Abstract

Time-Sensitive Networking (TSN) serves as a one-size-fits-all solution for mixed-criticality communication, in which flow scheduling is vital to guarantee real-time transmissions. Traditional approaches statically assign priorities to flows based on their associated applications, resulting in significant queuing delays. In this paper, we observe that assigning different priorities to a flow leads to varying delays due to different shaping mechanisms applied to different flow types. Leveraging this insight, we introduce a new scheduling method in mixed-criticality TSN that incorporates a priority adjustment scheme among diverse flow types to mitigate queuing delays and enhance schedulability. Specifically, we propose dependency-aware priority adjustment algorithms tailored to different link-overlapping conditions. Experiments in various settings validate the effectiveness of the proposed method, which enhances the schedulability by 20.57% compared with the SOTA method.

Exploiting Dependency-Aware Priority Adjustment for Mixed-Criticality TSN Flow Scheduling

TL;DR

This paper tackles the queuing delays that arise in mixed-criticality TSN when using static priorities. It introduces a dependency-aware adaptive priority adjustment method that operates across TT, AVB-A, and AVB-B flows, accounting for different shaping mechanisms and link overlaps. Two algorithmic regimes are proposed: Parallel-FG, where routes are disjoint, and Cross-FG, where FG conflicts are considered via TTConfliction and FG-conflict graphs with metrics TTAdjust and AVBAdjust; these yield schedules that maximize the number of flows meeting their deadlines under RTA-based analysis. Experimental results on realistic SAE, AFDX, and Ladder topologies show schedulability improvements up to about 20–29% over state-of-the-art baselines, along with higher link utilization and substantially lower computation times, demonstrating practical gains for real-time TSN systems.

Abstract

Time-Sensitive Networking (TSN) serves as a one-size-fits-all solution for mixed-criticality communication, in which flow scheduling is vital to guarantee real-time transmissions. Traditional approaches statically assign priorities to flows based on their associated applications, resulting in significant queuing delays. In this paper, we observe that assigning different priorities to a flow leads to varying delays due to different shaping mechanisms applied to different flow types. Leveraging this insight, we introduce a new scheduling method in mixed-criticality TSN that incorporates a priority adjustment scheme among diverse flow types to mitigate queuing delays and enhance schedulability. Specifically, we propose dependency-aware priority adjustment algorithms tailored to different link-overlapping conditions. Experiments in various settings validate the effectiveness of the proposed method, which enhances the schedulability by 20.57% compared with the SOTA method.
Paper Structure (18 sections, 7 equations, 8 figures, 2 algorithms)

This paper contains 18 sections, 7 equations, 8 figures, 2 algorithms.

Table of Contents

  1. Introduction
  2. System Model
  3. Network Model
  4. Flow Model
  5. Switch Model
  6. Problem Formulation
  7. Optimization Strategy
  8. Parallel-FG case
  9. Cross-FG case
  10. Evaluation
  11. Experimental setup
  12. Baseline
  13. Network setting
  14. Experimental Results
  15. Performance in different traffic volumes
  16. ...and 3 more sections

Figures (8)

  • Figure 1: Comparison between (a) Static priority assignment and (b) Adaptive priority adjustment. In (a), AVB-A flows experience severe queuing delay and eventually reach a timeout state. After adaptive priority adjustment in (b), the delays of tightly time-constrained flows are reduced. All flows are scheduled.
  • Figure 2: The TSN switch model supporting preemption mode. Express Mac (eMac) and preemptable Mac (pMac) are two Mac types defined in the TSN preemption specification. Flows in eMac can preempt flows in pMac.
  • Figure 3: Parallel FGs and crossed FGs. Flow $f_{Ri,j}^{x}$ means the $j$th flow of class $x$ in $FG_{Ri}$.
  • Figure 4: Scheduling process of AVB-A and AVB-B adjustment. The priority adjustment determines the priority for all the AVB flows, which forms the feasible solution.
  • Figure 5: Illustration of FG-conflict graph. (a) The network topology and flows. $f_{Ri,j}^{x}$ means the $j$th flow of class $x$ in $FG_{Ri}$. (b) The FG-conflict graph of (a). $u_{i,j}$ is the edge weight between $FG_{Ri}$ and $FG_{Rj}$. $f_{Ri}$ means the flow in $FG_{Ri}$.
  • ...and 3 more figures