Based on the release notes, Parallel Compression (and parallel filters) is supported as of 1.10.2 and later.
What is the recommended way to check for this capability via configure or CMake. For example, I have a library that uses HDF5 and I would like to enable the use of parallel compression if the underlying HDF5 library is recent enough to support the feature – What feature test should I use in MY CMake and autotools scripts to detect and enable the use of HDF5’s parallel compression.
Is there a symbol that was added at that time, or is it simply a version >= 1.10.2 test?
You should be able to look at the libhdf5.settings file in the lib/ directory of your HDF5 installation to determine whether parallel compression was enabled or not.
There is a feature called “Parallel Filtered Dataset Writes” in the libhdf5.settings file which may indicate parallel compression, but it is not clear. I entered bug HDFFV-10702 to add/improve the information in the libhdf5.settings file with regards to parallel compression. We will add this to the upcoming HDF5-1.10.5 release.
The HDF Group uses Jira to record bugs, issues, and feature requests. To access the bug report in Jira you must log in to jira.hdfgroup.org. If you don’t have an account, you can register for one here: https://www.hdfgroup.org/register/
