The NVIDIA cuSOLVER library provides a collection of dense and sparse direct solvers which deliver significant acceleration for Computer Vision, CFD, Computational Chemistry, and Linear Optimization applications.

     

Key Features

    
  • cusolverDN: Key LAPACK dense solvers 3-6x faster than MKL.
    • Dense Cholesky, LU, SVD, QR
    • Applications include: optimization, Computer Vision, CFD
  • cusolverSP
    • Sparse direct solvers & Eigensolvers
    • Applications include: Newton's method, Chemical Kinetics
  • cusolverRF
    • Sparse refactorization solver
    • Applications include: Chemistry, ODEs, Circuit simulation

Performance

Review the latest CUDA performance report to learn how much you could accelerate your code.

cuSOLVER 7.0, MKL 11.0.4, SuiteSparse 3.6.0, Acclerator Tesla K40, CPU i7-3930K @3.20 GHz

cuSOLVER 7.0, MKL 11.0.4, SuiteSparse 3.6.0, Acclerator Tesla K40, CPU i7-3930K @3.20 GHz

Availability

The cuSOLVER library is freely available as part of the CUDA 7 RC which currently is available exclusively to CUDA Registered Developers.