Bug located in ...

[mcMunich]

Before submitting the issue

• I have checked for Compatibility issues
• I have searched among the existing issues
• I am using a Python virtual environment

Description of the bug

The compression of the h5 migration does not have any effect on the size of the resulting .h5 file.

Steps To Reproduce

`
import os
from ansys.dpf import core as dpf
dpf.set_default_server_context(dpf.AvailableServerContexts.premium)
dpf.start_local_server(ansys_path=r"C:\Program Files\ANSYS Inc\v252") # Change path to Ansys Installation folder

cwd = r"D:\PYANSYS\dpf"
rst_file_path = r"D:\PYANSYS\dpf\file.rst"
h5_file_1_comp = os.path.join(cwd, "file_1comp.h5")
h5_file_8_comp = os.path.join(cwd, "file_8comp.h5")

dataSource_rst = dpf.DataSources(rst_file_path)

model = dpf.Model(dataSource_rst)

Migration

migrate_op = dpf.operators.result.migrate_to_h5dpf()
migrate_op.connect(-2, 1) # GZIP Level
migrate_op.inputs.export_floats.connect(False) # True if Conversion to float
migrate_op.inputs.filename.connect(h5_file_1_comp )
migrate_op.connect(1, "U")
migrate_op.inputs.all_time_sets.connect(True)
migrate_op.inputs.data_sources.connect(dataSource_rst)
migrate_op.run()

migrate_op = dpf.operators.result.migrate_to_h5dpf()
migrate_op.connect(-2, 8) # GZIP Level
migrate_op.inputs.export_floats.connect(False) # True if Conversion to float
migrate_op.inputs.filename.connect(h5_file_8_comp )
migrate_op.connect(1, "U")
migrate_op.inputs.all_time_sets.connect(True)
migrate_op.inputs.data_sources.connect(dataSource_rst)
migrate_op.run()

`

Which Operating System causes the issue?

Windows

Which DPF/Ansys version are you using?

Ansys 2025 R2

Which Python version causes the issue?

3.13

Installed packages

most recent version

[PProfizi]

@oparreno FYI

[oparreno]

@mcMunich,
Please check the dataset called "data" from one of the big fields in the h5dpf. If compression has been properly used, you should see a filter stating GZIP and the compression level as in the screenshot below:

In case you see an uncompressed dataset, this might come from a dataset size lower than the minimum compression threshold. You can modify the lower threshold through pin -5 as stated in the migrate to h5dpf operator documentation.

Please also note that typically we recommend using GZIP/ZSTD compression with lower level as we noticed it provides pretty good gains and increasing the level is sometimes resulting in just slightly better results but is much most costly in terms of performance.