HDF5 hyperslab
An HDF5 hyperslab is a selection of elements in a multidimensional dataspace expressed through rectangular blocks and their positions.
Selection geometry: a regular hyperslab uses a starting offset, stride, block count and block size for each dimension; selection operations can combine regions. [1]
Storage distinction: a hyperslab identifies logical elements, not disk allocation or a chunk layout. Selecting a small region can still require reading and decompressing complete intersecting chunks. [2]
Virtual datasets: source and destination selections determine which values appear at which logical coordinates. Filenames alone do not describe the mapping. [3]
Relations
- related
- hdf5, hdf5-virtual-dataset, hdf5-chunked-layout
Citations
- The HDF Group: H5Sselect_hyperslab and dataspace selections
- h5py 3.15.0: slicing and chunked storage
- The HDF Group: H5Pset_virtual mappings
Published by Hesela team · Updated
Report a correction · Editorial standards · Reuse guidance
Cite this page
Hesela team. HDF5 hyperslab. Hesela. 2026-10-18. https://hesela.dev/terms/hdf5-hyperslab/