PhD Position in Mesh Partitioning for Exascale Computing Systems
Develop novel optimization algorithms for parallel mesh partitioning within the Horizon Europe FEniCS Lighthouse project.
Best for:Best for computer science graduates with strong C++ skills, a background in high-performance computing, and an interest in graph partitioning or numerical solvers.
Overview
Inria Lyon is offering a fully funded 3-year PhD position focused on mesh partitioning for high-performance solvers on exascale computing systems. Operating within the Horizon Europe FEniCS Lighthouse project, the researcher will work alongside international academic and research partners to design optimization models that accurately reflect modern hierarchical HPC communication overheads. The candidate is expected to develop, analyze, and implement algorithms, integrate them into the open-source FEniCS framework, and contribute actively to scientific publications and project deliverables. The successful applicant will join the ROMA research team in Villeurbanne, France, starting in February 2027.
What you'll do
- Design, analyze, and implement novel algorithms for graph, hypergraph, and mesh partitioning targeting exascale systems.
- Collaborate closely with international partners in the FEniCS Lighthouse project to integrate developed partitioning solutions into the open-source FEniCS framework.
- Evaluate partitioner performance on modern supercomputing architectures featuring complex hierarchical topologies.
- Draft technical reports, project documentation, and scientific papers for publication in leading venues.
- Engage in multi-institutional and interdisciplinary teamwork across academic and laboratory environments.
Requirements
Degree required: Bac + 5 or equivalent Master's degree in Computer Science, Applied Mathematics, or a related field
Degree offered: PhD
Language: English; clear technical writing and presentation skills
Instruction: English
- Proficiency in C++ and Python programming languages
- Understanding of finite element methods (FEM) for solving partial differential equations
- Ability to develop and analyze novel algorithms for meshes, graphs, or hypergraphs
- Familiarity with modern high-performance computing (HPC) architectures including distributed memory, shared memory, or GPUs
- Open to all nationalities meeting the degree and administrative requirements of Inria and French regulations
Funding & terms
How to apply
Deadline: 2026-11-19
Applications must be submitted online directly through the Inria website before the deadline. Processing of applications sent through other channels is not guaranteed.
- CV
- Transcripts
- Cover letter
Contact
Uçar Bora· PhD Supervisor
Create a free profile to unlock your personal match score, the full requirements, funding details and how to apply — in English or Persian.