Fabien Geyer

A performance study of Audio Video Bridging in aeronautic Ethernet networks

Audio-Video-Bridging (AVB) is a promising new commercially available Ethernet-based standard providing mechanism for audio and video transmission over Ethernet supporting demanding audio and video transmission delay requirements. This paper addresses the applicability of using AVB in a fully-switched Ethernet network that covers safety-related functions in the aeronautics. Avionic systems leveraging such digital networks have stringent requirements in terms of audio quality, latency, and jitter; e.g., latency can be at most a few milliseconds. The result of this work is a performance study of audio transmission approaches in aeronautics where we address AVB without synchronization and AVB synchronization with 802.1AS. We pay special attention to the failure mode of losing synchronization during operation. Two real-world scenarios are addressed in the experimental results - a demonstration of a switched Ethernet aircraft cabin and a demonstration for a highly safety-related audio system. For these two scenarios hard performance bounds are required in terms of synchronous playback that cannot easily be fulfilled, especially when time synchronization is lost.

Deterministic OpenFlow: Performance evaluation of SDN hardware for avionic networks

Due to special requirements avionic networking devices are typically quite expensive. One way to reduce costs is to make use of commercial off the shelf devices and configure them in a way that gives similar performance. In this paper we evaluate the use of OpenFlow in the avionics environment in terms of performance and configuration. The main feature of OpenFlow is fine-grained access to the switchs forwarding plane. While it was primarily designed to offer high configurability and reduction of cost through harmonization of interfaces, in newer versions OpenFlow added support for traffic policing. In OpenFlow this is realized with meters that allow for quality of service enforcement on a hardware level as well as an arbitrary mapping of meters to flows. This paper shows how to make use of OpenFlows meter commands to achieve deterministic behavior and discusses its advantages and shortcomings. We then implement the proposed solution on a commercial off the shelf OpenFlow switch and compare the switching performance to a state of the art avionics switch used in current aircraft.

Network engineering for real-time networks: comparison of automotive and aeronautic industries approaches

With the advent of electronic x-by-wire applications with strict reliability and safety requirements, formal verification of safety-critical networks has become an important step of the design process in the automotive and aeronautic industries. We first review in this article the different network technologies and architectures used by both industries. We then present and compare the two prevailing mathematical frameworks used by each industry for validating the correct behavior of a network: schedulability analysis and network calculus. Via an empirical evaluation of both methods in two different use cases, we show the strengths and weaknesses of both methods, and derive a simple guideline on which method to use depending on the type of network used.

Generalizing Network Calculus Analysis to Derive Performance Guarantees for Multicast Flows

Guaranteeing performance bounds of data flows is an essential part of network engineering and certification of networks with real-time constraints. A prevalent analytical method to derive guarantees for end-to-end delay and bu↵er size is Deterministic Network Calculus (DNC). Due to the DNC system model, one decisive restriction is that only unicast flows can be analyzed. Previous attempts to analyze networks with multicast flows circumvented this restriction instead of overcoming it. E.g., they replaced the system model with an overly-pessimistic one that consists of unicast flows only. Such approaches impair modeling accuracy and thus inevitably result in inaccurate performance bounds. In this paper, we approach the problem of multicast flows di↵erently. We start from existing DNC analyses and generalize them to handle multicast flows. We contribute a novel analysis procedure that leaves the network model unaltered, preserves its accuracy, allows for DNC principles such as pay multiplexing only once, and therefore derives more accurate performance bounds than existing approaches.

TSimNet: An Industrial Time Sensitive Networking Simulation Framework Based on OMNeT++

The upcoming IEEE Time Sensitive Networking standard will offer a set of new functionality for Ethernet like frame preemption and frame replication that need careful evaluation. The IETF DetNet working group also works on deterministic layer 3 segments and leverages therefore works from the IEEE group. We present here a simulation framework based on OMNeT++ that implements all non time-based features present in TSN in a modular and easily extendable way. The framework gives other researches a stable base to start evaluating existing and future scenarios. An evaluation of frame preemption as well as per-stream policing based on the framework is presented. It was shown, that frame preemption does not always bring latency advantages and highly depends on configuration of the system. The paper concludes with an overview of the computational cost for the use of this framework.

Towards stochastic flow-level network modeling: Performance evaluation of short TCP flows

We present in this paper a stochastic flow-level network model for the performance evaluation of IP networks with multiple bottlenecks supporting short-lived and long-lived TCP flows. Flow-level network models are efficient at estimating the mean bandwidth of TCP flows in various topologies, but they are generally limited to the study of infinite flows. This paper extends such models in order to evaluate short-lived flows alternating between idle and active periods where data of random size is transferred. We study the interaction between multiple flows and derive mean bandwidths, durations of file transfer or average number of active flows. We first study a single bottleneck, and then extend our analysis to networks with multiple bottlenecks as well as the effect of slow-start. We apply our results to various networks and assess the accuracy of our approach by comparing our analytical results with results of the discrete event simulator ns-2.

Evaluation of Audio/Video Bridging forwarding method in an avionics switched ethernet context

Possible evolution of Avionics Full-Duplex Switched Ethernet (AFDX) are currently investigated. Such network should support existing time-constrained avionics flows, as well as new best effort traffic, requiring scheduling algorithms to meet guarantees on avionics flows. Following the current trend of the automotive industry to move toward Audio/Video Bridging (AVB), this paper evaluates an application of the scheduling scheme described in the emerging IEEE 802.1 AVB standard in an avionics switched Ethernet scenario. Results are compared to other common scheduling strategies such as strict priority queuing, and different fair scheduling variants.

Performance Evaluation of Network Topologies using Graph-Based Deep Learning

Understanding the performance of network protocols and communication networks generally relies on expert knowledge and understanding of the different elements of a network, their configuration and the overall architecture and topology. Machine learning is often proposed as a tool to help modeling complex protocols. One drawback of this method is that high-level features are generally used - which require expert knowledge on the network protocols to be chosen, correctly engineered, and measured -- and the approaches are generally limited to a given network topology. In this paper, we propose a methodology to address the challenge of working with machine learning by using lower-level features, namely only a description of the network architecture. Our main contribution is an approach for applying deep learning on network topologies via the use of Graph Gated Neural Networks, a specialized recurrent neural network for graphs. Our approach enables us to make performance predictions based only on a graph-based representation of network topologies. We apply our approach to the task of predicting the throughput of TCP flows. We evaluate three different traffic models: large file transfers, small file transfers, and a combination of small and large file transfers. Numerical results show that our approach is able to learn the throughput performance of TCP flows with good accuracies larger than 90%, even on larger topologies.

Self-configuring real-time communication network based on OpenFlow

One of the major tasks when deploying real-time Ethernet networks is their configuration to achieve real-time behavior. In this paper we present an approach for a self-configuring plug-n-play network that automatically sets up devices and offers hard real-time guarantees to such devices. A new architecture and a protocol based on OpenFlow are proposed to achieve a system that can react to failures of switches and links by trying to repair such failures on a network level and restoring a full redundancy level while maintaining hard real-time guarantees. Implementation details are explained and the architecture is evaluated against randomly generated topologies. It is shown, that the solution can achieve a seamless failure recovery in link failures without any complexity at the end-device side.

Deterministic Network Calculus Analysis of Multicast Flows

Guaranteeing performance bounds of data flows is an essential part of network engineering and certification of networks with real-time constraints. A prevalent analytical method to derive guarantees for end-to-end delay and buffer size is Deterministic Network Calculus (DNC). Due to the DNC system model, one decisive restriction is that only unicast flows can be analyzed. Previous attempts to analyze networks with multicast flows circumvented this restriction instead of overcoming it. E.g., they replaced the system model with an overly-pessimistic one that consists of unicast flows only. Such approaches impair modeling accuracy and thus inevitably result in inaccurate performance bounds.