Enhancing Performance Bounds of Multiple-Ring Networks with Cyclic Dependencies based on Network Calculus

TL;DR

This paper introduces the PMOC concept to tighten delay bounds in cyclic ring networks using Network Calculus, improving resource efficiency and scalability, with validation through analysis and an avionics case study.

Contribution

The paper proposes the PMOC approach to reduce delay bound pessimism in cyclic networks, enhancing performance analysis accuracy and network resource utilization.

Findings

PMOC significantly tightens delay bounds compared to traditional methods.

The approach improves network resource efficiency and scalability.

Validation confirms effectiveness in a realistic avionics scenario.

Abstract

Tightening performance bounds of ring networks with cyclic dependencies is still an open problem in the literature. In this paper, we tackle such a challenging issue based on Network Calculus. First, we review the conventional timing approaches in the area and identify their main limitations, in terms of delay bounds pessimism. Afterwards, we have introduced a new concept called Pay Multiplexing Only at Convergence points (PMOC) to overcome such limitations. PMOC considers the flow serialization phenomena along the flow path, by paying the bursts of interfering flows only at the convergence points. The guaranteed endto- end service curves under such a concept have been defined and proved for mono-ring and multiple-ring networks, as well as under Arbitrary and Fixed Priority multiplexing. A sensitivity analysis of the computed delay bounds for mono and multiple-ring networks is conducted with respect to various flow and network parameters, and their tightness is assessed in comparison with an achievable worst-case delay. A noticeable enhancement of the delay bounds, thus network resource efficiency and scalability, is highlighted under our proposal with reference to conventional approaches. Finally, the efficiency of the PMOC approach to provide timing guarantees is confirmed in the case of a realistic avionics application.