PARALUTION is a library that enables you to perform various sparse iterative solvers and preconditioners on multi/many-core CPU and GPU devices. Based on C++, it provides a generic and flexible design that allows seamless integration with other scientific software packages.

Advantages

1

Features

PARALUTION contains Krylov subspace solvers (CR, CG, BiCGStab, GMRES, IDR), Multigrid (GMG, AMG), Deflated PCG, Fixed-point iteration schemes, Mixed-precision schemes and fine-grained parallel preconditioners based on splitting, ILU factorization with levels, multi-elimination ILU factorization, additive Schwarz and approximate inverse. The library also provides iterative eigenvalue solvers.

1

Multi-platform Support

The library can be compiled under Linux/Unix-like , Windows and Mac OS. PARALUTION provides multi-core CPU/Host (OpenMP), NVIDIA GPU (CUDA, OpenCL), AMD GPU (OpenCL), Intel Xeon Phi/MIC (OpenCL, OpenMP/offload mode) support, including VS (Visual Studio) gcc (GNU C++) and icc (Intel C++) compilers.

1

Dual License Model

The PARALUTION library is released under dual license model – GPLv3 and commercial. All documents are under Creative Commons licenses. Please visit our License page for more details.

Latest News

10 Reasons to Try PARALUTION!

  • Open Source / Dual License Model
  • Easy to use
  • No special hardware/library required
  • Generic and robust design
  • Various iterative solvers/preconditioners
  • Portable code and results
  • Supports many sparse matrix formats
  • Extendable backends (new hardware)
  • Seamless integration
  • PARALUTION IS FUN!