Skip to content

tensorstore sharding #141

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 2 commits into from
Apr 8, 2025
Merged

tensorstore sharding #141

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e88dc73
COMP: Update pixi.lock for 0.45.0
thewtex Apr 8, 2025
c6a8589
ENH: Tensorstore sharding support
thewtex Apr 8, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion .github/workflows/pixi-test.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,7 +15,7 @@ jobs:

- uses: prefix-dev/setup-pixi@v0.8.1
with:
pixi-version: v0.44.0
pixi-version: v0.45.0
frozen: true

- name: Run tests
Expand Down
76 changes: 57 additions & 19 deletions ngff_zarr/to_ngff_zarr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -108,7 +108,13 @@ def _numpy_to_zarr_dtype(dtype):


def _write_with_tensorstore(
store_path: str, array, region, chunks, zarr_format, dimension_names=None
store_path: str,
array,
region,
chunks,
zarr_format,
dimension_names=None,
internal_chunk_shape=None,
) -> None:
"""Write array using tensorstore backend"""
import tensorstore as ts
Expand Down Expand Up @@ -136,6 +142,13 @@ def _write_with_tensorstore(
spec["metadata"]["data_type"] = _numpy_to_zarr_dtype(array.dtype)
if dimension_names:
spec["metadata"]["dimension_names"] = dimension_names
if internal_chunk_shape:
spec["metadata"]["codecs"] = [
{
"name": "sharding_indexed",
"configuration": {"chunk_shape": internal_chunk_shape},
}
]
else:
raise ValueError(f"Unsupported zarr format: {zarr_format}")
dataset = ts.open(spec, create=True, dtype=array.dtype).result()
Expand Down Expand Up @@ -319,6 +332,7 @@ def to_ngff_zarr(
sharding_kwargs["codecs"] = previous_codecs + [sharding_codec]
else:
sharding_kwargs["codecs"] = [sharding_codec]
internal_chunk_shape = c0
arr = arr.rechunk(shards)

if memory_usage(image) > config.memory_target and multiscales.scale_factors:
Expand Down Expand Up @@ -469,15 +483,27 @@ def to_ngff_zarr(
)
if use_tensorstore:
scale_path = f"{store_path}/{path}"
_write_with_tensorstore(
scale_path,
optimized,
region,
[c[0] for c in arr_region.chunks],
zarr_format=zarr_format,
dimension_names=dimension_names,
**kwargs,
)
if chunks_per_shard is None:
_write_with_tensorstore(
scale_path,
optimized,
region,
[c[0] for c in arr_region.chunks],
zarr_format=zarr_format,
dimension_names=dimension_names,
**kwargs,
)
else:
_write_with_tensorstore(
scale_path,
optimized,
region,
shards,
zarr_format=zarr_format,
dimension_names=dimension_names,
internal_chunk_shape=internal_chunk_shape,
**kwargs,
)
else:
dask.array.to_zarr(
optimized,
Expand All @@ -501,15 +527,27 @@ def to_ngff_zarr(
if use_tensorstore:
scale_path = f"{store_path}/{path}"
region = tuple([slice(arr.shape[i]) for i in range(arr.ndim)])
_write_with_tensorstore(
scale_path,
arr,
region,
[c[0] for c in arr.chunks],
zarr_format=zarr_format,
dimension_names=dimension_names,
**kwargs,
)
if chunks_per_shard is None:
_write_with_tensorstore(
scale_path,
arr,
region,
[c[0] for c in arr.chunks],
zarr_format=zarr_format,
dimension_names=dimension_names,
**kwargs,
)
else: # Sharding
_write_with_tensorstore(
scale_path,
arr,
region,
shards,
zarr_format=zarr_format,
dimension_names=dimension_names,
internal_chunk_shape=internal_chunk_shape,
**kwargs,
)
else:
arr = _prep_for_to_zarr(store, arr)
dask.array.to_zarr(
Expand Down
Loading
Loading