Projects

High-performance computing

  • Poseidon: Performance Optimization For StEncIl-baseD OceaN Simulations (2023-2025)
    • Short description: This project will work on bridging a permanently increasing gap between the numerics and HPC with a separation of concerns: A so-called domain-specific language (DSL) will be used in which the applied mathematicians can express their model equations so that an HPC expert can develop tools to automatically transform this into highly performing code on the target programming model and corresponding architecture. As part of this, we will use an already existing intermediate representation (IR) with PsyIR (part of the PsyClone development) which already allows parsing code in the Fortran programming language so that the developers of numerical methods can stick to the language they are familiar with.
      To ensure a long-lasting impact, this project will primarily focus on the ocean simulation code CROCO but also target future applicability to the NEMO model.
  • Crocodiles: Inria associate team (2023-2026)
    • Short description: In the domain of High-Performance Computing (HPC) optimization for ocean simulation codes, HPC optimization of ocean models requires skills and a deeper understanding of HPC from all the hardware and software layers. This associate team is sharing experiences and collaborating on such developments with ANL and their Flash-X code developments.
    • Collaborators: Anshy Dubey et al from ANL.
  • A Comprehensive Software Stack for Dynamic Resources Management: JLESC team (2023-)
    • Link with further information
    • Short description: Dynamic Resource Management (DRM) allows for dynamic changes of the resources assigned to a job during its execution. DRM has gained considerable interest over the last years as it could provide many benefits to providers of HPC systems and their users, such as improving energy efficiency and throughput.
      However, DRM requires changes and interactions throughout the whole HPC software stack. This project aims to integrate the promising Dynamic Processes with PSets (DPP) approach for managing MPI ranks with the Dynamic Management of Resources (DMR) programming interface.
      Once provided with the DRM framework, several scientific applications and benchmarks will be updated with the dynamic resources paradigm and evaluated regarding coding usability and reconfiguration performance.
    • Collaborators: BSC & DataMove
  • SEANERGYS (2024-)
    • t.b.d.

Les commentaires sont clos.