Tuesday 5thWednesday 6thThursday 7thFriday 8th
8:30-9:00First session Opening remarks
9:00-10:00First session cont. Keynote
Securing Cyber-Physical Systems with Varying Levels of Autonomy: Can we Win, or at Least Not Lose?
Miroslav Pajic
Real-Time High Performance Applications: the ARM approach
Matteo Andreozzi
10:00-10:30Coffee breakCoffee breakCoffee breakCoffee break
10:30-12:00Second session Session #1: network and real-time calculus Session #3: scheduling and response-time analysis Session #6: cache coherence
(starting at 11:00)
13:30-15:00Third session Session #2: machine learning for real-time systems Session #4: applications Session #7: outstanding papers
15:00-15:30Coffee breakCoffee breakCoffee breakCoffee break
15:30-17:00Fourth session Industrial challenge session (see description on discord) Session #5: mixed-criticality scheduling Retrospective on ECRTS
Gerhard Fohler
& Closing remarks
17:00-19:00First-timer receptionInteractive session (real-time pitches, journal2conference, artifacts) and receptionSocial event

Keynote speech: Securing Cyber-Physical Systems with Varying Levels of Autonomy: Can we Win, or at Least Not Lose?

Miroslav Pajic is the Dickinson Family Associate Professor in the Department of Electrical and Computer Engineering at Duke University. He also holds a secondary appointment in the Computer Science Department. His research interests focus on the design and analysis of high-assurance cyber-physical systems with varying levels of autonomy and human interaction, at the intersection of (more traditional) areas of embedded systems, AI, learning and controls, formal methods and robotics. He received various awards including the NSF CAREER Award, ONR Young Investigator Program  Award, ACM SIGBED Early-Career Researcher Award, IEEE TCCPS Early-Career Award, IBM Faculty Award, ACM SIGBED Frank Anger Memorial Award, the Joseph and Rosaline Wolf Dissertation Award from Penn Engineering, as well as seven Best Paper and Runner-up Awards, such as the Best Paper Awards at the 2017 ACM SIGBED International Conference on Embedded Software (EMSOFT) and 2014 ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS), and the Best Student Paper award at the 2012 IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS). He is an associate editor in the ACM Transactions on Cyber-Physical Systems and ACM Transactions Computing for Healthcare (ACM HEALTH), and was a co-Chair of the 2019 ACM/IEEE International Conference on Cyber-Physical Systems (ICCPS’19).

Keynote speech: Real-Time High Performance Applications: the ARM approach

Abstract: Computer Systems are rapidly evolving and moving from being designed as targeting general purpose, real-time, high performance etc. use cases, to be designed for increasingly mixed and often difficult to predict use-cases. Co-location of multiple workloads on a single computer system allows improving system’s resources utilization, re-use (e.g. IO devices, accelerators, etc.) and efficiency of data sharing across workloads. This, however, comes at the cost of potential severe performance degradation due to interference on shared resources, and increased uncertainty in terms of workload performance predictability. I present here ARM’s approach to solving the challenges that high-performance applications pose, showing how understanding predictability bottlenecks is crucial to this challenge, and how to adopt analysis strategies suitable for this class of applications. I proceed to describe current ARM IP features that can increase predictability and determinism by implementing QoS and I finally present our vision for future ARM QoS solutions. 

Bio: Matteo Andreozzi contributes at ARM to the design of future architectures for mobile, automotive, embedded and robotic systems. He looks after themes such as QoS, realtime and performance of ARM based systems. Matteo leads a team that is responsible for designing architectural solutions for high-performance real-time systems, covering all major ARM market segments: mobile and IoT devices, infrastructure, automotive and industrial. His team’s activities are varied and span from research to prototyping, partner engagement, technology transfer and  ARM architecture design and deployment. His experience with topics such as event-driven simulation, QoS, RealTime , goes back almost 15 year ago, when he started his academic researches and Ph.D. and developed into industrial roles first at Nvidia, and now ARM. He has been an invited lecturer at both the University of Pisa and at the University of Cambridge where he gave talks about the topics above.

Session #1: network and real-time calculus (session chair: Georg von der Brüggen)

  • Unikernel-Based Real-Time Virtualization under Deferrable Servers: Analysis and Realization
    Kuan-Hsun Chen (University of Twente); Mario Günzel (TU Dortmund University); Boguslaw Jablkowski, Markus Buschhoff (EMVICORE GmbH); Jian-Jia Chen (TU Dortmund University)
  • A Mathematical Comparison Between Response-Time Analysis and Real-Time Calculus for Fixed-Priority Preemptive Scheduling
    Victor Pollex (INCHRON AG); Frank Slomka (Ulm University)
  • A Formal Link between Response Time Analysis and Network Calculus
    Marc Boyer (ONERA); Sophie Quinton (INRIA Grenoble); Pierre Roux (ONERA)

Session #2: machine learning for real-time systems (session chair: Renato Mancuso)

  • RTScale: Sensitivity-aware Adaptive Image Scaling for Real-time Object Detection
    Seonyeong Heo (ETH Zurich); Shinnung Jeong, Hanjun Kim (Yonsei University)
  • Predictable programming framework for machine learning applications in safety-critical systems
    Iryna De Albuquerque Silva (ANITI/ONERA); Thomas Carle (IRIT – Univ Toulouse 3 – CNRS); Adrien Gauffriau (Airbus); Claire Pagetti (ONERA)
  • Using Quantile Regression in Neural Networks for Contention Prediction in Multicore Processors
    Axel Brado (Barcelona Supercomputing Center); Isabel Serra (Barcelona Supercomputing Center and Centre de Recerca Matemàtica); Enrico Mezzetti (Barcelona Supercomputing Center and Maspatechnologies); Jaume Abella (Barcelona Supercomputing Center and Maspatechnologies); Francisco J Cazorla (Barcelona Supercomputing Center and Maspatechnologies)

Session #3: scheduling and response-time analysis (session chair: Björn Brandenburg)

  • Scheduling offset-free systems under FIFO priority protocol
    Matheus Ladeira (ISAE-ENSMA, Université de Poitiers – Poitiers, France), Emmanuel Grolleau (ISAE-ENSMA, Université de Poitiers – Poitiers, France); Fabien Bonneval, Gautier Hattenberger (ENAC, Université de Toulouse – Toulouse, France); Yassine Ouhammou (ISAE-ENSMA, Université de Poitiers – Poitiers, France), Yuri Hérouard (ISAE-ENSMA, Université de Poitiers – Poitiers, France)
  • Response-Time Analysis for Non-Preemptive Periodic Moldable Gang Tasks
    Geoffrey Nelissen (Eindhoven University of Technology), Joan Marcè I Igual (Eindhoven University of Technology), Mitra Nasri (Eindhoven University of Technology)
  • Response-Time Analysis for Self-Suspending Tasks Under EDF Scheduling
    Federico Aromolo (Scuola Superiore Sant’Anna), Alessandro Biondi (Scuola Superiore Sant’Anna), Geoffrey Nelissen (Eindhoven University of Technology)

Session #4: applications (session chair: Marko Bertogna)

  • General framework for Routing, Scheduling and Formal Timing Analysis in Deterministic Time-aware Networks
    Anais Finzi (TTTech Computertechnik AG), Ramon Serna Oliver (TTTech Computertechnik AG)
  • Correctness and Efficiency Criteria for the Multi-Phase Task Model
    Rémi Meunier (AUSY – IRIT – INSA); Thomas Carle (IRIT – Université Toulouse 3 – CNRS); Thierry Monteil (IRIT – CNRS – INSA)
  • Overrun-Resilient Multiprocessor Real-Time Locking
    Zelin Tong (UNC Chapel Hill), Shareef Ahmed (UNC Chapel Hill), Jim Anderson (UNC Chapel Hill)

Session #5: mixed-criticality scheduling (session chair: Kuan-Hsun Chen)

  • An Approach to Formally Specifying the Behaviour of Mixed-Criticality Systems
    Alan Burns (University of York); Cliff Jones (University of Newcastle)
  • Achieving Isolation in Mixed-Criticality Industrial Edge Systems with Real-Time Containers
    Marco Barletta (Università degli Studi di Napoli Federico II), Marcello Cinque (Università degli Studi di Napoli Federico II), Luigi De Simone (Università degli Studi di Napoli Federico II), Raffaele Della Corte (Università degli Studi di Napoli Federico II)

Session #6: cache coherence (session chair: Francisco Cazorla)

  • Parallelism-Aware High-Performance Cache Coherence with Tight Latency Bounds
    Reza Mirosanlou (University of Waterloo); Mohamed Hassan (McMaster University); Rodolfo Pellizzoni (University of Waterloo)
  • Predictably and Efficiently Integrating COTS Cache Coherence in Real-Time Systems
    Mohamed Hossam (McMaster University), Mohamed Hassan (McMaster University)

Session #7: outstanding papers (session chair: Martina Maggio)

  • RT-DFI: Optimizing Data-Flow Integrity for Real-Time Systems
    Nicolas Bellec (Univ Rennes, Inria, CNRS, IRISA); Guillaume Hiet (CentraleSupélec); Simon Rokicki (Univ Rennes, Inria, CNRS, IRISA); Frédéric Tronel (CentraleSupélec); Isabelle Puaut (Univ Rennes, Inria, CNRS, IRISA)
  • Foundational Response-Time Analysis as Explainable Evidence of Timeliness
    Marco Maida (Max Planck Institute for Software Systems MPI-SWS), Sergey Bozhko (Max Planck Institute for Software Systems MPI-SWS), Björn Brandenburg (Max Planck Institute for Software Systems MPI-SWS)
  • Using Markov’s Inequality with Power-of-k Function for Probabilistic WCET Estimation
    Sergi Vilardell (Barcelona Supercomputer Center and Universitat Politècnica de Catalunya); Isabel Serra (Centre de Recerca Matemàtica); Enrico Mezzetti (Barcelona Supercomputing Center and Maspatechnologies), Jaume Abella (Barcelona Supercomputing Center and Maspatechnologies); Joan del Castillo (Universitat Autònoma de Barcelona); Francisco J. Cazorla (Barcelona Supercomputing Center and Maspatechnologies)

Comments are closed.