https://www.mdu.se/

Modern GPUs offer high computational throughput but are not designed to ensure spatial or temporal isolation between concurrent tasks. Only a few server-grade devices with multi-instance GPU technology support true hardware partitioning. As consumer and embedded GPUs now include many Streaming Multiprocessors (SMs), running multiple tasks concurrently has become common to improve utilization. Softwarebased SM-level spatial partitioning, where each partition is restricted to a fixed set of SMs, offers a simple and portable way to isolate concurrent tasks. Although in this method SMs are isolated, resources such as the L2 cache and DRAM remain shared, causing memory interference that can affect execution times and lead to timing unpredictability, often overlooked in real-time analysis. This paper empirically characterizes cross-partition interference on a COTS GPU under software-only SM-level partitioning. Using a mix of compute- and memory-intensive tasks, partitioned and non-partitioned configurations are compared by measuring both end-to-end response times and GPU execution times. The results indicate that SM-level partitioning largely confines interference for compute-bound tasks. However, memory-bound tasks continue to cause substantial interference across partitions, increasing the task execution time by as much as twofold. These findings indicate that memory interference is a critical challenge for achieving timing predictability in real-time GPU-based cyber-physical systems, underscoring the need to address memory interference to achieve predictable performance.

Time-Sensitive Networking (TSN) enables deterministic mixed-criticality communication over standard Ethernet through time-aware scheduling (IEEE 802.1Qbv) for Scheduled Traffic (ST) and via Credit-Based Shaping (CBS) for Audio-Video Bridging (AVB) traffic. Existing scheduling methods often prioritize ST schedules and require iterative feedback loops between ST-schedule generation and AVB schedulability analysis to ensure the timing of both classes. We propose a novel approach called EAST (Efficient AVB-aware Scheduling for TSN) that improves runtime and schedulability by introducing a constraint-driven approach that decouples AVB analysis from ST schedule synthesis. We first perform a Worst-Case Response Time (WCRT) analysis to determine the maximum ST interference each AVB stream can tolerate without missing its deadline. This interference bound is then translated into temporal constraints on the ST scheduler, ensuring that both ST and AVB traffic remain schedulable without requiring multiple iterations of generating ST schedules and checking AVB schedulability. The resulting formulation guarantees end-to-end latency bounds for both ST and AVB streams while drastically reducing schedule synthesis time. Synthetic and real-world test cases show that EAST reduces runtime by several orders of magnitude compared to prior methods, while maintaining or improving overall schedulability.

The convergence of OPC UA Publish–Subscribe (PubSub) and Time-Sensitive Networking (TSN) enables vendor-agnostic and predictable communication in real-time distributed industrial automation systems. Deploying such systems, however, requires consistent configuration across application and network layers that currently expose mismatched timing abstractions. TSN depends on precise traffic specifications, while OPC UA PubSub provides extensive configuration options without guidance on selecting timing-related parameters, creating engineering complexity and increasing commissioning effort. To address this challenge, this paper presents TPConf, a tool-supported framework that co-configures OPC UA PubSub and TSN from traffic specifications by classifying application requirements into industrial automation traffic classes, mapping them to TSN classes (ST, AVB, BE), and deriving corresponding PubSub parameters. TPConf further generates ready-to-compile publisher code for open62541, and an industrial use case with diverse simulated flows demonstrates the automated co-configuration workflow. By unifying application and network configuration, TPConf reduces engineering effort and supports predictable and scalable OPC UA over TSN deployments in real-time automation systems.

Interoperability across industrial automation systems is a cornerstone of Industry 4.0. To address this need, the OPC Unified Architecture (OPC UA) Publish-Subscribe (PubSub) model offers a promising mechanism for enabling efficient communication among heterogeneous devices. PubSub facilitates resource sharing and communication configuration between devices, but it lacks clear guidelines for mapping diverse industrial traffic types to appropriate PubSub configurations. This gap can lead to misconfigurations that degrade network performance and compromise real-time requirements. This paper proposes a set of guidelines for mapping industrial traffic types, based on their timing and quality-of-service specifications, to OPC UA PubSub configurations. The goal is to ensure predictable communication and support real-time performance in industrial networks. The proposed guidelines are evaluated through an industrial use case that demonstrates the impact of incorrect configuration on latency and throughput. The results underline the importance of traffic-aware PubSub configuration for achieving interoperability in Industry 4.0 systems.

This paper identifies that existing industrial automation standards, such as IEC/IEEE 60802 and IEEE 802.1Q, often have inconsistent definitions of traffic types. In the context of utilizing time-sensitive networking (TSN) standards for future automation systems, clear and consistent traffic characteristics and use cases should be defined to benefit from TSN features. Besides that, to facilitate the integration of TSN into the automation systems, the current standards provide a recommendation for mapping the automation traffic to the TSN traffic classes. In this paper, we propose an alternative mapping methodology for automation traffic to TSN traffic classes after presenting the existing automation traffic and their characteristics. Finally, through a case study, we show the potential of the new mapping methodology compared to the standard mapping strategy.

Formal verification of robotic applications, particularly those based on ROS 2, is desirable for ensuring correctness and safety. However, the complexity of formal methods and the manual effort required for model creation and parameter extraction often hinder their adoption. This paper addresses these challenges by proposing a model-based methodology that automates the formal verification process using model-driven engineering techniques. We introduce a methodology which can be applied as a toolchain that automates the initialization of formal model templates in UPPAAL using system parameters derived from ROS 2 execution traces generated by the ROS2_tracing tool. The toolchain employs four model representations based on custom Eclipse Ecore metamodels to capture both structural and verification aspects of ROS 2 systems. The methodology supports both implemented and conceptual systems and enables iterative verification of timing and scheduling parameters through model-to-model and model-to-text transformations. A proof-of-concept implementation demonstrates the feasibility of the proposed approach. The designed toolchain supports verification using two types of UPPAAL models: one for individual node verification (e.g., callback latency and buffer overflow) and another for end-to-end latency analysis of ROS 2 processing chains. Experiments conducted on two implemented and one conceptual ROS 2 systems validate the correctness and adaptability of the toolchain. The results show that the toolchain can automate parameter extraction and model generation. The proposed methodology modularizes the verification process, allowing domain experts to focus on their areas of expertise. It targets to enhances traceability and reusability across different verification scenarios and formal models. The approach aims to make formal verification more accessible and practical to robotics developers.

Integrating Time-Sensitive Networking (TSN) with 5G cellular networks facilitates high-bandwidth, low-latency end-to-end communication in networked embedded systems. An integrated TSN-5G network has the potential to support predictable and deterministic end-to-end communication, as well as to significantly enhance scalability, particularly in industrial automation, by providing flexibility, efficiency, and responsiveness. This paper aims to facilitate the end-to-end (E2E) communication over the integrated TSN-5G networks, by addressing two of the main challenges: (1) design and implementation of an effective TSN-5G gateway that not only ensures an effective forwarding mechanism to translate the traffic among both networks, but also maps the TSN traffic to the corresponding 5G quality-of-service profiles to ensure the timing requirements of highly critical traffic, and (2) establishment of clock synchronization across all network components to support the E2E communication in TSN-5G networks. The paper outlines the design and implementation of a robust TSN-5G gateway that bridges TSN and 5G network architectures, ensuring seamless interoperability between them. We utilize a standalone private 5G network within a controlled lab environment to establish the E2E communication for the TSN-5G network, with a particular emphasizes on analyzing latencies and jitter to provide valuable insights for industrial implementation. Moreover, the gateway facilitates a time synchronization approach, to enable time coordination between network components that supports E2E communication on TSN-5G networks considering the hardware limitation. Our findings indicate that achieving latencies below 20 ms is feasible in an integrated TSN-5G network using the proposed configuration of a private 5G setup with a channel bandwidth of 40 MHz.

Time-Sensitive Networking (TSN) has emerged as a key technology for ensuring reliable and deterministic communication in vehicular networks. However, the complexity of evaluating, configuring, and analyzing TSN mechanisms has hindered its widespread adoption. This paper introduces CONAN-TSN, an integrated toolchain designed to address these challenges by providing a comprehensive solution for the configuration, analysis, and evaluation of TSN networks. CONAN-TSN includes tools for traffic generation, traffic mapping, scheduling, and timing analysis, enabling seamless integration and practical implementation of TSN in vehicular systems. The toolchain’s effectiveness is demonstrated through its application to an industrial use case, showcasing its ability to enhance network schedulability and performance. The results highlight CONAN-TSN’s potential to bridge the gap between theoretical benefits and real-world deployment of TSN, offering a valuable resource for researchers and practitioners in the field.

A hybrid cloud is an efficient solution to deal with the problem of insufficient resources of a private cloud when computing demands increase beyond its resource capacities. Cost-efficient workflow scheduling, considering security requirements and data dependency among tasks, is a prominent issue in the hybrid cloud. To address this problem, we propose a mathematical model that minimizes the monetary cost of executing a workflow and satisfies the security requirements of tasks under a deadline. The proposed model fulfills data dependency among tasks, and data transmission time is formulated with exact mathematical expressions. The derived model is a Mixed-integer linear programming problem. We evaluate the proposed model with real-world workflows over changes in the input variables of the model, such as the deadline and security requirements. This paper also presents a post-optimality analysis that investigates the stability of the assignment problem. The experimental results show that the proposed model minimizes the cost by decreasing inter-cloud communications for dependent tasks. However, the optimal solutions are affected by the limitations that are imposed by the problem constraints. 

The increasing complexity of modern embedded systems highlights the limitations of Controller Area Network (CAN) in terms of transmission speed and scalability. The IEEE Time-Sensitive Networking (TSN) task group developed a set of standards to enhance switched Ethernet with high bandwidth, low jitter, and deterministic communication. Despite these advances, CAN will likely co-exist with TSN in, e.g., the automotive industry due to factors such as cost-effectiveness and legacy of CAN. This paper presents an experimental evaluation of a CAN-toTSN gateway implementation, focusing on the impact of different forwarding and scheduling strategies on network performance. We analyze various queuing techniques and scheduling mechanisms in a realistic experimental setup and assess their impact on end-to-end delay and TSN bandwidth utilization. The evaluation results demonstrate that encapsulating only a single CAN frame within a TSN frame effectively minimizes the end-to-end delay of CAN frames, in particular when a high-speed TSN network is used. Furthermore, we perform a comparative evaluation of the Time-Aware Shaper (TAS) and Weighted Round Robin (WRR) mechanisms in the TSN network. Interestingly, WRR leads to lower delays for CAN frames in the TSN network compared to TAS, which we attribute to the lack of synchronization between CAN and TSN.