Skip to content

Recompress all chunks independent of size #17

@jbusecke

Description

@jbusecke

From reading the man page here:

All  rechunked variables are de-interlaced with the HDF5 shuffle filter
       (which significantly improves compression) before being compressed with
       zlib (see the -z option), and also have the  Fletcher32  HDF5  checksum
       algorithm activated.

Does this mean if a file with large chunks is provided (lets say 200MB), nothing will be done to the compression? Or am I misunderstanding things here?

I was just talking to folks in the QC/QA working group, and mentioned that consistent compression across dataset files is a hard requirement for our efforts to produce virtual zarr references to the data and it seems to me if the application of this tool is required for publication it would present an excellent opportunity to harmonize compression across all files.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions