SolverStack
SolverStack@Inria Bordeaux
SolverStack aims at providing a coherent, high-performance (HPC) linear algebra solver stack. It provides a comprehensive collection of numerical solvers, partitioning tools, and runtime systems designed for modern supercomputers operating on dense and sparse matrices.
The ecosystem includes direct, iterative (Krylov), and hybrid direct/iterative methods, with advanced preconditioners and low-rank compression techniques, ensuring portability of performance from multicore laptops to petascale and exascale supercomputers.
👉 Official Website: https://solverstack.gitlabpages.inria.fr/
Software Ecosystem
- Sparse Direct Methods: PaStiX, qr_mumps
- Dense Direct Methods: Chameleon
- Sparse Matrix Package: SPM
- Iterative & Krylov Methods: Fabulous
- Hybrid Direct/Iterative Solvers: Composyx
- Randomized & Low-Rank Decompositions: FMR
- Task-Based Runtime Systems: StarPU
- Communication Libraries: NewMadeleine
- Partitioning & Ordering: Scotch
Packaging & Distribution
SolverStack software components are systematically packaged and maintained for high-performance computing environments:
- Guix-HPC:
guix install <package>(guix-hpc) - Spack:
spack install <package>(spack) - Homebrew:
brew install <package>(brew-repo)
Partner Teams & Institutions
SolverStack software is developed by researchers and engineers in joint project-teams at Inria Bordeaux Sud-Ouest, Bordeaux University, CNRS (LaBRI UMR 5800), and Bordeaux INP:
Contact: solverstack@inria.fr
