• Unified memory management for CPUs and GPUs with extended smart pointers.
• BLAS-like functions enclosed to templated vectors and expression templates for simple and efficient vector operations.
• Flexible parallel reduction and parallel scan for GPUs and multicore CPUs.
• Parallel sorting including inplace sorting with user defined swapping.
• Various types of matrices like dense, tridiagonal, multidiagonal and general sparse matrices based on number of different formats.
• Lambda matrices  with matrix elements  defined by C++ lambda functions.
• Segments - data abstraction of sparse matrix formats like CSR, Adaptive CSR, Ellpack, Sliced Ellpack, Chunked Ellpack, Bisection Ellpack.
• Linear solvers - CG, BiCGStab,GMRES, parallel CWYGMRES, TFQMR, Jacobi, SOR.
• Preconditioners of linear solvers - Jacobi, ILU0 (CPU only), ILUT( CPU only).
• ODE solvers - Euler (first order), Runge-Kutta-Merson (fourth order adaptive).
• Orthogonal numerical grids.
• Unstructured numerical meshes (including distributed) with import from Netgen, VTU, VTK, XMLVTK, PVTU and PVTK.
• Image formats - DICOM (via DCMTK), JPEG (via libjpeg), PNG (via libpng), PGM.

TNL is not a library of one or a few numerical methods. We are interested in implementing new methods and data structures which may profit from templated design and support of modern hardware. If you feel it the same way you are very welcome to ...