dask-array

Dask array with query optimization

Motivation

Dask Array is powerful, but requires expertise to drive effectively.

The dask-array project reimplements dask.array, but represents your calculation at a level where it can be optimized intelligently before executed. This allows the project to reorder and replace calculations to provide the same result but with a more efficient path.

Installation

pip install dask-array

Usage

This project looks and feels like Dask Array

import dask_array as da

x = da.ones((1000, 1000), chunks=(100, 100))

y = x + x.T
result = y[:100, :100]

But when you go to compute, your calculation gets rewritten to be more efficient. This is apparent if you look at the query structure of the underlying array.

>>> result.pprint()
Operation             Shape    Bytes   Chunks
  Getitem          (100, 100)   78 kiB  100×100
  └ Add          (1000, 1000)  7.6 MiB  100×100
    ├ Ones       (1000, 1000)  7.6 MiB  100×100
    └ Transpose  (1000, 1000)  7.6 MiB  100×100
      └ Ones     (1000, 1000)  7.6 MiB  100×100

This calculation starts from a large array expression, then takes a small slice. It is more efficient to push that slice into the expression before building the task graph.

The optimize function rewrites things automatically.

>>> result.optimize().pprint()
Operation            Shape   Bytes   Chunks
  FusedBlockwise  (100, 100)  78 kiB  100×100
  └ Ones          (100, 100)  78 kiB  100×100

You don't need to call optimize though. Dask compute/persist machinery will do this for you. You just need to change your import.

# import dask.array as da
import dask_array as da

Native Frisky acceleration

The normal Dask scheduler path is pure Python and works without a Rust toolchain. On platforms with a native wheel, dask-array also includes a Rust accelerator that Frisky can use to submit compact task records instead of materializing large Python task graphs. This is automatic when computing with a Frisky scheduler; otherwise computations stay on the standard Dask path.

Xarray

Dask-array can replace the default dask.array by registering itself as an Xarray chunk manager.

from dask_array.xarray import register
register()

After this all xarray calculations will benefit from query optimization.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

dask_array-0.3.0.tar.gz (498.6 kB view details)

Uploaded Jun 25, 2026 Source

Built Distributions

If you're not sure about the file name format, learn more about wheel file names.

The dropdown lists show the available interpreters, ABIs, and platforms. Enable javascript to be able to filter the list of wheel files.

dask_array-0.3.0-py3-none-any.whl (359.2 kB view details)

Uploaded Jun 25, 2026 Python 3

dask_array-0.3.0-cp310-abi3-win_amd64.whl (763.6 kB view details)

Uploaded Jun 25, 2026 CPython 3.10+Windows x86-64

dask_array-0.3.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (907.8 kB view details)

Uploaded Jun 25, 2026 CPython 3.10+manylinux: glibc 2.17+ x86-64

dask_array-0.3.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl (901.0 kB view details)

Uploaded Jun 25, 2026 CPython 3.10+manylinux: glibc 2.17+ ARM64

dask_array-0.3.0-cp310-abi3-macosx_11_0_arm64.whl (861.9 kB view details)

Uploaded Jun 25, 2026 CPython 3.10+macOS 11.0+ ARM64

File details

Details for the file dask_array-0.3.0.tar.gz.

File metadata

Download URL: dask_array-0.3.0.tar.gz
Upload date: Jun 25, 2026
Size: 498.6 kB
Tags: Source
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for dask_array-0.3.0.tar.gz
Algorithm	Hash digest
SHA256	`3835e34e81575a09dfcb1cfec9d33f56e29a1ea0431aed23849212eb711d095c`
MD5	`013fb221eafb0bf951808c1d01dbc95e`
BLAKE2b-256	`ee333d19a32c47513a63a9d8a8e8e84458f8f57ea85667b5b4babbe97597eeac`

See more details on using hashes here.

Provenance

The following attestation bundles were made for dask_array-0.3.0.tar.gz:

Publisher: publish.yml on mrocklin/dask-array

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: dask_array-0.3.0.tar.gz
- Subject digest: 3835e34e81575a09dfcb1cfec9d33f56e29a1ea0431aed23849212eb711d095c
- Sigstore transparency entry: 1946938614
- Sigstore integration time: Jun 25, 2026
Source repository:
- Permalink: mrocklin/dask-array@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Branch / Tag: refs/tags/0.3.0
- Owner: https://github.com/mrocklin
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Trigger Event: push

File details

Details for the file dask_array-0.3.0-py3-none-any.whl.

File metadata

Download URL: dask_array-0.3.0-py3-none-any.whl
Upload date: Jun 25, 2026
Size: 359.2 kB
Tags: Python 3
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for dask_array-0.3.0-py3-none-any.whl
Algorithm	Hash digest
SHA256	`4acd8df03576ce2255476b1a80494328acf29630783cf8daa05f4860fa3d4c67`
MD5	`405aa68492829d8d6c2e548a633b2f80`
BLAKE2b-256	`4658986909ac2760f89134f5035461fd5805e2e872b439f02e00ec760f75e5d2`

See more details on using hashes here.

Provenance

The following attestation bundles were made for dask_array-0.3.0-py3-none-any.whl:

Publisher: publish.yml on mrocklin/dask-array

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: dask_array-0.3.0-py3-none-any.whl
- Subject digest: 4acd8df03576ce2255476b1a80494328acf29630783cf8daa05f4860fa3d4c67
- Sigstore transparency entry: 1946938883
- Sigstore integration time: Jun 25, 2026
Source repository:
- Permalink: mrocklin/dask-array@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Branch / Tag: refs/tags/0.3.0
- Owner: https://github.com/mrocklin
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Trigger Event: push

File details

Details for the file dask_array-0.3.0-cp310-abi3-win_amd64.whl.

File metadata

Download URL: dask_array-0.3.0-cp310-abi3-win_amd64.whl
Upload date: Jun 25, 2026
Size: 763.6 kB
Tags: CPython 3.10+, Windows x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for dask_array-0.3.0-cp310-abi3-win_amd64.whl
Algorithm	Hash digest
SHA256	`43578c49bb61dcf10f1978de0bdf546e0dce57bb59c7409354245d005416f68f`
MD5	`9e88fdc5415f9084a4434ca05ca66bff`
BLAKE2b-256	`4428c849824604105a2d5118c81dc5ef1a38273c1f31321c3430b82ff09d1764`

See more details on using hashes here.

Provenance

The following attestation bundles were made for dask_array-0.3.0-cp310-abi3-win_amd64.whl:

Publisher: publish.yml on mrocklin/dask-array

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: dask_array-0.3.0-cp310-abi3-win_amd64.whl
- Subject digest: 43578c49bb61dcf10f1978de0bdf546e0dce57bb59c7409354245d005416f68f
- Sigstore transparency entry: 1946939005
- Sigstore integration time: Jun 25, 2026
Source repository:
- Permalink: mrocklin/dask-array@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Branch / Tag: refs/tags/0.3.0
- Owner: https://github.com/mrocklin
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Trigger Event: push

File details

Details for the file dask_array-0.3.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.

File metadata

Download URL: dask_array-0.3.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Upload date: Jun 25, 2026
Size: 907.8 kB
Tags: CPython 3.10+, manylinux: glibc 2.17+ x86-64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for dask_array-0.3.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
Algorithm	Hash digest
SHA256	`e127e4e396b54c59c31e9166aa56bb570289e03835eae3bc8eaf0ab567ccfd88`
MD5	`07a982e5baaea0798457269b3d5c2a2b`
BLAKE2b-256	`e7e8d57f278f245420fbab372c25abe703a818aca98a322e9eee3c35bb5daccb`

See more details on using hashes here.

Provenance

The following attestation bundles were made for dask_array-0.3.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl:

Publisher: publish.yml on mrocklin/dask-array

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: dask_array-0.3.0-cp310-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl
- Subject digest: e127e4e396b54c59c31e9166aa56bb570289e03835eae3bc8eaf0ab567ccfd88
- Sigstore transparency entry: 1946939255
- Sigstore integration time: Jun 25, 2026
Source repository:
- Permalink: mrocklin/dask-array@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Branch / Tag: refs/tags/0.3.0
- Owner: https://github.com/mrocklin
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Trigger Event: push

File details

Details for the file dask_array-0.3.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl.

File metadata

Download URL: dask_array-0.3.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Upload date: Jun 25, 2026
Size: 901.0 kB
Tags: CPython 3.10+, manylinux: glibc 2.17+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for dask_array-0.3.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
Algorithm	Hash digest
SHA256	`4135d6457e09473a3a6b1f53a4821b84dd0a68c3d6235145245a7b45ab52ee0f`
MD5	`18e0425ebaccabb08942647887f2864b`
BLAKE2b-256	`0eca41574c739f130aee3796e33fea249d3011df7cae3f0580b102313243c5d6`

See more details on using hashes here.

Provenance

The following attestation bundles were made for dask_array-0.3.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl:

Publisher: publish.yml on mrocklin/dask-array

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: dask_array-0.3.0-cp310-abi3-manylinux_2_17_aarch64.manylinux2014_aarch64.whl
- Subject digest: 4135d6457e09473a3a6b1f53a4821b84dd0a68c3d6235145245a7b45ab52ee0f
- Sigstore transparency entry: 1946939376
- Sigstore integration time: Jun 25, 2026
Source repository:
- Permalink: mrocklin/dask-array@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Branch / Tag: refs/tags/0.3.0
- Owner: https://github.com/mrocklin
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Trigger Event: push

File details

Details for the file dask_array-0.3.0-cp310-abi3-macosx_11_0_arm64.whl.

File metadata

Download URL: dask_array-0.3.0-cp310-abi3-macosx_11_0_arm64.whl
Upload date: Jun 25, 2026
Size: 861.9 kB
Tags: CPython 3.10+, macOS 11.0+ ARM64
Uploaded using Trusted Publishing? Yes
Uploaded via: twine/6.1.0 CPython/3.13.12

File hashes

Hashes for dask_array-0.3.0-cp310-abi3-macosx_11_0_arm64.whl
Algorithm	Hash digest
SHA256	`10e7a309bbb494d80eea23dc927ace295df8f1b5199ed6ed2715da1ccc41501c`
MD5	`d4389d4d06a3cd051c6409b1f6846fda`
BLAKE2b-256	`0b544ed1180c7c9e1016da1ae2d36d5a5637372df1588072a51a7a6dacad55aa`

See more details on using hashes here.

Provenance

The following attestation bundles were made for dask_array-0.3.0-cp310-abi3-macosx_11_0_arm64.whl:

Publisher: publish.yml on mrocklin/dask-array

Attestations: Values shown here reflect the state when the release was signed and may no longer be current.

Statement:
- Statement type: https://in-toto.io/Statement/v1
- Predicate type: https://docs.pypi.org/attestations/publish/v1
- Subject name: dask_array-0.3.0-cp310-abi3-macosx_11_0_arm64.whl
- Subject digest: 10e7a309bbb494d80eea23dc927ace295df8f1b5199ed6ed2715da1ccc41501c
- Sigstore transparency entry: 1946939140
- Sigstore integration time: Jun 25, 2026
Source repository:
- Permalink: mrocklin/dask-array@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Branch / Tag: refs/tags/0.3.0
- Owner: https://github.com/mrocklin
- Access: public
Publication detail:
- Token Issuer: https://token.actions.githubusercontent.com
- Runner Environment: github-hosted
- Publication workflow: publish.yml@fc8e3d777661c234ef9725ca911c4c5bf70d80b1
- Trigger Event: push

Release history Release notifications | RSS feed

0.7.1

Jul 22, 2026

0.7.0

Jul 20, 2026

0.6.0

Jul 6, 2026

0.5.0

Jul 4, 2026

0.4.3

Jun 30, 2026

0.4.2

Jun 30, 2026

0.4.1

Jun 28, 2026

0.4.0

Jun 27, 2026

This release

0.3.0

Jun 25, 2026

0.2.0

Jun 23, 2026

0.1.0

Jun 9, 2026

dask-array 0.3.0

dask-array

Motivation

Installation

Usage

Native Frisky acceleration

Xarray

Project links

Key dates

1 maintainer

GitHub Statistics

License expression

Requires

Provides Extra

Download files

Source Distribution

Built Distributions

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

File details

File metadata

File hashes

Provenance

Release history Release notifications | RSS feed