RTAS is a top-tier conference with a focus on systems with timing requirements. RTAS’25 welcomes papers describing applications, case studies, methodologies, tools, algorithms or operating systems, middleware or hardware innovations that contribute to the state of the art in the design, implementation, validation, verification, and evolution of systems with timing requirements. RTAS’25 consists of two tracks:

    Track 1. Systems and Applications;
    Track 2. Applied Methodologies and Foundations.

Scope and Guidelines

To be in scope, papers must explicitly consider at least one of the following:

    Some kind of timing requirements;
    Improvements or innovations that directly support the fundamental properties of systems with timing requirements.

Timing requirements

The timing requirements of interest are broadly defined and include not only classical hard real-time constraints, but also soft real-time, probabilistic, quality-of-service (QoS), throughput or latency requirements. As an example, work that makes an AI algorithm run faster is not in scope, but work that provides guarantees on the response time of an AI algorithm is in scope. The authors should specifically state the types of timing-related properties addressed by the contribution(s) presented in their paper.

Improvements or innovations to support the fundamental properties of systems with timing requirements

Work that supports the design, implementation, validation, or evaluation of deterministic, predictable, dependable, efficient, or next-generation systems with timing requirements are welcome. The authors should specifically state how their work can be used or built upon to ultimately achieve systems with timing requirements beyond the state of the arts. For example, work on verification and automated test case generation of the codebase of an operating system to achieve timing or timing isolation guarantees is in scope. As another example, work on a compiler that reduces the WCET or provides sound distributions of timing information and timing variability of code is in scope. In contrast, a paper on automated test case generation for functional correctness of embedded systems is not in scope, as it does not provide improvements to support systems with timing requirements.

The application area can be any type of systems with timing requirements, including but not limited to: resource-constrained embedded systems, distributed cyber-physical systems (CPS), cloud/edge/fog computing systems, cloud data centers, Internet of Things (IoT), mobile computing, robotics, smart grid, and smart cities, as well as middleware and frameworks, machine learning and signal processing algorithms. RTAS welcomes both papers backed by formal proofs, as well as papers that focus exclusively on empirical validation of timing requirements, e.g., using traces or performance models inferred from data. Research results from fundamental research, (case-driven) applied research, and (pragmatic) industry practice are all in scope.

RTAS’25 follows a double-anonymous peer reviewing process: author identities and affiliations will not be revealed to reviewers. Authors will have the opportunity to provide a response to reviews before acceptance decisions are made, solely to provide clarifications and correct misconceptions. The response will not allow authors to introduce new material beyond the original submission or promise such material for the camera-ready version. There will be an optional artifact evaluation process for accepted papers that assesses the reproducibility of the work.

Track 1: Systems and Applications

This track focuses on research of an empirical nature pertaining to (system- or component-) level analysis, optimization, and verification, as well as applications, runtime software, and hardware architectures for systems with timing requirements. Topics relevant to this track include, but are not limited to:

    applications with timing requirements,
    real-time and embedded operating systems,
    hypervisors and runtime frameworks,
    hardware architectures, memory hierarchies, FPGAs, GPUs and accelerators,
    networks with timing requirements,
    CPS/IoT infrastructure,
    microservice technologies, cloud and edge computing, real-time artificial intelligence and machine learning,
    application profiling, WCET analysis, compilers, tools, benchmarks and case studies.

Papers discussing design and implementation experiences on real industrial systems are especially encouraged. Papers submitted to this track should focus on specific systems and implementations. Authors must include a section with experimental results performed on a real implementation or demonstrate applicability to an industrial case study or working system. The experiment or case study discussions must highlight the key lessons learned. Simulation-based results are acceptable for architectural simulation, or other cases where authors clearly motivate why it is not feasible to develop and evaluate a real system.

Empirical survey-based research focused on the real-time systems field is also welcome in this track. This type of research uses surveys, questionnaires, interviews, use cases or other empirical techniques to obtain information about the past / current / future state of play in the research, design, development, verification, validation, and deployment of systems with timing requirements.

Track 2: Applied Methodologies and Foundations

This track focuses on fundamental models and analysis techniques/methods applicable to systems with timing requirements to solve problems. The track welcomes knowledge-based models, models built from data, as well as a combination, and different types of analysis methods, including analytical, statistical, or probabilistic methods. Topics relevant to this track include, but are not limited to:

    modelling languages, modelling methods, model learning, model validation and calibration,
    scheduling and resource allocation,
    system-level optimization and co-design techniques,
    design space exploration,
    verification and validation methodologies.

Papers must describe the main context or use case for the proposed methods giving clear motivating examples based on real systems. The system models and any assumptions used in the derivation of the methods must be applicable to real systems and reflect actual needs. Papers must include a section on experimental results, preferably including a case study based on information from a real system. The use of synthetic workloads and models is acceptable if appropriately motivated and used to provide a systematic evaluation.