EN FR
EN FR


Section: Partnerships and Cooperations

European Initiatives

FP7 & H2020 Projects

ARGO

Participants : Damien Hardy, Isabelle Puaut, Stefanos Skalistis.

  • Title: Argo: WCET-Aware Parallelization of Model-Based Applications for Heterogeneous Parallel Systems

  • Program: H2020

  • Type: RIA

  • Duration: Jan 2016 – Mar 2019

  • Coordinator: Karlsruhe Institut für Technologie (Germany)

  • Université de Rennes 1 contact: Steven Derrien

  • Partners:

    • Karlsruher Institut für Technologie (Germany)

    • SCILAB enterprises SAS (France)

    • Université de Rennes 1 (France)

    • Technologiko Ekpaideftiko Idryma (TEI) Dytikis Elladas (Greece)

    • Absint GmbH (Germany)

    • Deutsches Zentrum für Luft- und Raumfahrt EV (Germany)

    • Fraunhofer (Germany)

  • Increasing performance and reducing costs, while maintaining safety levels and programmability are the key demands for embedded and cyber-physical systems in European domains, e.g. aerospace, automation, and automotive. For many applications, the necessary performance with low energy consumption can only be provided by customized computing platforms based on heterogeneous many-core architectures. However, their parallel programming with time-critical embedded applications suffers from a complex toolchain and programming process. Argo (WCET-Aware PaRallelization of Model-Based Applications for HeteroGeneOus Parallel Systems) will address this challenge with a holistic approach for programming heterogeneous multi- and many-core architectures using automatic parallelization of model-based real-time applications. Argo will enhance WCET-aware automatic parallelization by a crosslayer programming approach combining automatic tool-based and user-guided parallelization to reduce the need for expertise in programming parallel heterogeneous architectures. The Argo approach will be assessed and demonstrated by prototyping comprehensive time-critical applications from both aerospace and industrial automation domains on customized heterogeneous many-core platforms.

Argo also involves Steven Derrien and Angeliki Kritikakou from the CAIRN team.

HiPEAC4 NoE

Participants : Pierre Michaud, Erven Rohou, André Seznec, Isabelle Puaut.

P. Michaud, A. Seznec and E. Rohou are members of the European Network of Excellence HiPEAC4.

HiPEAC4 addresses the design and implementation of high-performance commodity computing devices in the 10+ year horizon, covering both the processor design, the optimizing compiler infrastructure, and the evaluation of upcoming applications made possible by the increased computing power of future devices.

Eurolab-4-HPC

Participant : Erven Rohou.

  • Title: EuroLab-4-HPC: Foundations of a European Research Center of Excellence in High Performance Computing Systems

  • Program: H2020

  • Duration: September 2018 – September 2020

  • Coordinator: Chalmers Tekniska Hoegskola AB (Sweden)

  • Partners:

    • Barcelona Supercomputing Center - Centro Nacional de Supercomputacion (Spain)

    • Chalmers Tekniska Hoegskola (Sweden)

    • Foundation for Research and Technology Hellas (Greece)

    • Universität Stuttgart (Germany)

    • The University of Manchester (United Kingdom)

    • Inria (France)

    • Universität Augsburg (Germany)

    • ETH Zürich (Switzerland)

    • École Polytechnique Federale de Lausanne (Switzerland)

    • Technion - Israel Institute of Technology (Israel)

    • The University of Edinburgh (United Kingdom)

    • Rheinisch-Westfaelische Technische Hochschule Aachen (Germany)

    • Universiteit Gent (Belgium)

  • Inria contact: Albert Cohen (Inria Paris)

  • Europe has built momentum in becoming a leader in large parts of the HPC ecosystem. It has brought together technical and business stakeholders from application developers via system software to exascale systems. Despite such gains, excellence in high performance computing systems is often fragmented and opportunities for synergy missed. To compete internationally, Europe must bring together the best research groups to tackle the long-term challenges for HPC. These typically cut across layers, e.g., performance, energy efficiency and dependability, so excellence in research must target all the layers in the system stack. The EuroLab-4-HPC project's bold overall goal is to build connected and sustainable leadership in high-performance computing systems by bringing together the different and leading performance oriented communities in Europe, working across all layers of the system stack and, at the same time, fueling new industries in HPC.