This article was accepted into the corpus but its outbound wikilinks were never NER-processed — typical at the deepest BFS hop or when the run's entity cap was reached. No expansion funnel to show.
| SLEPc | |
|---|---|
| Name | SLEPc |
| Operating system | Cross-platform |
| Platform | POSIX, Windows |
| Genre | Numerical linear algebra, eigenvalue solvers |
SLEPc is a software library for the solution of large-scale sparse eigenvalue and related spectral problems on parallel computers. It builds on abstractions for distributed linear algebra and parallel computation to provide scalable solvers, preconditioners, and spectral transformations for applications in science and engineering.
SLEPc is designed for high-performance computations in fields that include computational fluid dynamics, structural mechanics, quantum chemistry, geophysics, and material science, serving projects associated with Argonne National Laboratory, Lawrence Livermore National Laboratory, Los Alamos National Laboratory, CERN, and universities such as Massachusetts Institute of Technology, Stanford University, University of Cambridge, University of Oxford. Its design complements libraries like PETSc, Trilinos, Hypre, and BLAS, enabling workflows used by researchers at institutions such as Princeton University, California Institute of Technology, ETH Zurich, Technical University of Munich, EPFL.
SLEPc provides modular implementations of algorithms used in computational projects spanning collaborations with teams at NASA, European Space Agency, National Institute of Standards and Technology, and industry partners including Siemens, General Electric, Schlumberger, IBM, Intel. The library supports dense and sparse operators studied in settings at Los Alamos National Laboratory, Oak Ridge National Laboratory, Rutherford Appleton Laboratory, and integrates techniques from work originating at Courant Institute, Kaiser Wilhelm Institute, Max Planck Society, and research groups such as those at Princeton Plasma Physics Laboratory. It implements Krylov methods, Jacobi–Davidson, subspace iteration, and contour integral methods used in projects resembling efforts by Microsoft Research and Google Research.
SLEPc's architecture relies on abstractions for distributed matrices and vectors compatible with PETSc and interoperable with packaging systems like Spack, Conda, Homebrew, and build tools such as CMake and Autotools. The design philosophy reflects parallel computing paradigms developed at National Center for Supercomputing Applications, Los Alamos National Laboratory, Argonne National Laboratory, and incorporates ideas from libraries like ScaLAPACK, Elemental, SUNDIALS, SuiteSparse, and FFTW. It targets execution environments including clusters funded by programs like Horizon 2020, NSF-supported facilities, and supercomputers such as Summit (supercomputer), Fugaku, Titan (supercomputer), Blue Gene/Q-class systems.
SLEPc installs on Unix-like platforms and Microsoft Windows through compatibility layers and package managers used in environments maintained by organizations including Red Hat, Canonical, SUSE, and distributions employed at Los Alamos National Laboratory, Jet Propulsion Laboratory, and universities like University of California, Berkeley. Compatibility matrices often reference compilers and toolchains from GCC, Clang, Intel C++ Compiler, and MPI implementations such as Open MPI, MPICH, and Intel MPI. Binary packaging and continuous integration workflows draw on services provided by Travis CI, Jenkins, GitHub Actions, and repositories like GitHub, GitLab, and mirrors hosted by SourceForge.
Typical workflows demonstrate eigenvalue computations for problems studied in collaborations with MIT Lincoln Laboratory, Bell Labs, Siemens PLM Software, and research groups at Imperial College London, University of Toronto, McGill University, and University of Illinois Urbana-Champaign. Example applications include modal analysis of structures encountered in projects with Boeing, Airbus, and Rolls-Royce, electronic structure calculations relevant to work at Lawrence Berkeley National Laboratory and Argonne National Laboratory, and stability analyses like those pursued at Princeton University and Stanford University. Tutorials and example code frequently reference numerical linear algebra tasks familiar to users of MATLAB, Octave, Python (programming language), Fortran, and C++) bindings.
Performance tuning for SLEPc targets large-scale simulations run on systems operated by NERSC, PRACE, XSEDE, and national labs including Oak Ridge National Laboratory and Argonne National Laboratory. Benchmarks compare solver performance with implementations in ARPACK, PARPACK, SLEPc-adjacent tools like Krylov-Schur and algorithms developed in collaborations similar to those at ETH Zurich, RWTH Aachen University, University of Michigan, Purdue University, and University of Texas at Austin. Scalability studies are relevant to exascale initiatives supported by DOE offices and partnerships with vendors such as NVIDIA, AMD, Intel, and system integrators like Cray Inc..
Development of SLEPc is coordinated by academic and national laboratory teams, with contributions from researchers affiliated with University of Valladolid, University of Bath, University of Santiago de Compostela, and community members collaborating via platforms like GitHub and GitLab. The project follows collaborative development practices similar to those used by Linux Foundation projects and research software governed under policies seen at European Commission-funded consortia and National Science Foundation initiatives. Licensing and governance models align with open-source ecosystems alongside projects such as GNU Project, Apache Software Foundation, and BSD licenses.