This Tucker-Octree (T-Octree) library compresses 3-dimensional dense datasets using octree multiresolution approach with Tucker decomposition. The collected core and factor matrices are further compressed with the ZFP library.
Typical cmake installation:
> mkdir build; cd build
> cmake .. -DCMAKE_BUILD_TYPE=Release -DEigen3_DIR="path/to/share/eigen3/cmake" -Dzfp_DIR="path/to/lib/cmake/zfp" -DCMAKE_INSTALL_PREFIX="<vlasiator-path>/vdf_compression"
> make installAfter that add -Lpath/to/vdf_compression/lib to linker flags and
-Ipath/to/vdf_compression/include to compiler flags for vlasiator.
T-Octree is licensed under GPL2.0 or later (see LICENSE text).
T-Octree utilizes the following codes:
- zfp (>=v1.0.1, may work with earlier)
- Eigen3 (>=v3.4.0, may work with earlier)
- Spectra C++ Library For Large Scale Eigenvalue Problems (included in
contrib/include) - argparse (included in
contrib/include)
This software library was built within project Adaptive Strategies Towards Expedient Recovery In eXascale (ASTERIX) at CSC – IT Center for Science Ltd. and University of Helsinki. Innovation Study ASTERIX has received funding through the Inno4scale project, which is funded by the European High-Performance Computing Joint Undertaking (JU) under Grant Agreement No 101118139. The JU receives support from the European Union's Horizon Europe Programme.