Remove titiler cmr benchmarking

[abarciauskas-bgse]

This PR removes:

• the datacube-benchmark package (see more notes on this below)
• API references for the datacube-benchmark package
• Notebooks describing the methodology for and reporting on TiTiler-CMR benchmarking and compatibility

The visualization section of the documentation is looking a little sparse and TiTiler-focused. Should we add content about the other methods of visualization (direct client libraries, python libraries) or at least link to that content in the Zarr Visualization Report?

datacube-benchmark package

This package contains both utilities for benchmarking TiTiler-CMR (datacube_benchmark.titiler) as well as more general-purpose synthetic datacube generation + testing utilities (datacube_benchmark.*). The former I think belongs in the TiTiler-CMR repo. The latter looks very useful! But I'm not sure where/when they were used. It seems like this code could be in another repo and then used for more general purpose datacube/zarr store benchmarking against various APIs?

[maxrjones]

thanks @abarciauskas-bgse! I agree with the CMR specific components going into the Titiler-CMR repository. However, the general tools were used for demonstrating the impact of the "worst practices" defined in the guide (e.g., https://developmentseed.org/datacube-guide/worst-practices/tiny-chunks.html#demonstrating-storage-inefficiencies-of-too-small-of-chunks). Could you please leave those and only remove the titiler specific components?

[abarciauskas-bgse]

Of course! My mistake for missing this documentation that depends on those datacube_benchmark utilities. I had contemplated separating the titiler-cmr utilities out from the zarr/chunk utilities previously but decided it wasn’t worth the effort if they are not being used. But I was wrong about them not being used!

…

On Fri, Dec 12, 2025 at 10:57 AM Max Jones ***@***.***> wrote: *maxrjones* left a comment (developmentseed/datacube-guide#16) <#16 (comment)> thanks @abarciauskas-bgse <https://github.com/abarciauskas-bgse>! I agree with the CMR specific components going into the Titiler-CMR repository. However, the general tools were used for demonstrating the impact of the "worst practices" defined in the guide (e.g., https://developmentseed.org/datacube-guide/worst-practices/tiny-chunks.html#demonstrating-storage-inefficiencies-of-too-small-of-chunks). Could you please leave those and only remove the titiler specific components? — Reply to this email directly, view it on GitHub <#16 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ADSSGTD3HWLQVZC5UEU5ZGT4BMFX7AVCNFSM6AAAAACOQXTXVCVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZTMNBXG42TCNBZGU> . You are receiving this because you were mentioned.Message ID: ***@***.***>

[abarciauskas-bgse]

@maxrjones after adding back the datacube_benchmark package (minus the titiler subdirectory) I tried to run some of the notebooks using the instructions in the README, i.e.:

uv run ipython kernel install --user --env VIRTUAL_ENV $(pwd)/.venv --name=project
uv run --with jupyter jupyter lab

but noticed these notebooks depend on some zarr stores in Azure, for which I got the following error:

Do you think we could make these stores public?

[maxrjones]

@maxrjones after adding back the datacube_benchmark package (minus the titiler subdirectory) I tried to run some of the notebooks using the instructions in the README, i.e.:

uv run ipython kernel install --user --env VIRTUAL_ENV $(pwd)/.venv --name=project
uv run --with jupyter jupyter lab

but noticed these notebooks depend on some zarr stores in Azure, for which I got the following error:

Do you think we could make these stores public?

Thanks for pointing this out. I'll address this separately.