Davis Bennett
Davis Bennett
I added some slice-based benchmarks based on the examples from https://github.com/zarr-developers/zarr-python/issues/3524, and I updated the contributing docs with a section about the benchmarks. assuming we can resolve the discussion about...
new problem: the codspeed CI benchmarks are way too slow! the benchmark suite runs in 90s locally, and It's taking over 40m to run in CI. Help would be appreciated...
owing to the large number of syscalls in our benchmark code, codspeed recommended using the [walltime instrument](https://codspeed.io/docs/instruments/walltime) instead of their virtual CPU instrument. But to turn on the walltime benchmark,...
looks like the walltime instrument is working! I think this is g2g
> IMO it'd be better to skip the tests/benchmarks during regular test runs in the interest of speed i think this makes sense -- on my workstation the current benchmark...
cc @kylebarron , @LDeakin
> vlen-bytes is the codec for that data type, but the data type name is variable_length_bytes. oops, good catch! I updated the table with this correction
thanks for this report! @mkitti do you have any ideas for what zarr python is doing wrong here?
> Is deferring the write to `zarr.json` until store closure a viable option? I think the problem you are having with `zarr.json` also holds for chunk files. We could cache...
thanks for that context, I can see how streaming data introduces the requirement to write `zarr.json` at the end. I wonder if the current design of the `Array` class is...