An efficient scheduling approach for multi-level industrial chain flows in time-sensitive networking

Abstract

Time-sensitive networking (TSN) is a promising real-time network technology that has deterministic delay and jitter guarantee capabilities. However, TSN flow scheduling faces challenges in the multi-level topology due to the cooperative scheduling requirements of multiple data flows, which are abstracted as TSN chain flows. In this paper, we present an efficient scheduling approach, namely TSN Chained Flow Scheduling (TCFS), which is utilized to globally schedule chain flows in the multi-level topology. To obtain optimal TCFS scheduling results, we design the scheduling constraints and propose an offline algorithm to efficiently schedule chain flows in the multi-level topology. Based on offline scheduling results, we further propose a heuristic method that includes time-tabling and sequencing algorithms to improve the dynamic scheduling performance. Experimental results show that the TCFS approach efficiently improves the schedulability for chain flows, as compared with benchmarks.