Skip to content

Conversation

@dcherian
Copy link
Contributor

Disclaimer Used Claude to generate the benchmark

Pad once instead of repeatedly in serial for every dimension being coarsened.

| Before [6a7a11b5] <main>   | After [873f7ce6] |   Ratio | Benchmark (Parameter)                        |
|----------------------------|------------------|---------|----------------------------------------------|
| 856M                       | 727M             |    0.85 | coarsen.Coarsen.peakmem_coarsen_with_padding |
| 59.7±1ms                   | 49.9±0.4ms       |    0.84 | coarsen.Coarsen.time_coarsen_with_padding    |
| 729M                       | 597M             |    0.82 | coarsen.Coarsen.peakmem_coarsen_no_padding   |
| 55.2±0.4ms                 | 39.5±0.1ms       |    0.71 | coarsen.Coarsen.time_coarsen_no_padding      |
  • Closes #xxxx
  • Tests added
  • User visible changes (including notable bug fixes) are documented in whats-new.rst
  • New functions/methods are listed in api.rst

Pad once instead of repeatedly in serial for every dimension being
coarsened.
```
| Before [6a7a11b] <main>   | After [873f7ce6] |   Ratio | Benchmark (Parameter)                        |
|----------------------------|------------------|---------|----------------------------------------------|
| 856M                       | 727M             |    0.85 | coarsen.Coarsen.peakmem_coarsen_with_padding |
| 59.7±1ms                   | 49.9±0.4ms       |    0.84 | coarsen.Coarsen.time_coarsen_with_padding    |
| 729M                       | 597M             |    0.82 | coarsen.Coarsen.peakmem_coarsen_no_padding   |
| 55.2±0.4ms                 | 39.5±0.1ms       |    0.71 | coarsen.Coarsen.time_coarsen_no_padding      |
```

Co-authored-by: Claude <[email protected]>
@dcherian dcherian added the run-benchmark Run the ASV benchmark workflow label Nov 14, 2025
@dcherian dcherian mentioned this pull request Nov 14, 2025
19 tasks
@dcherian dcherian enabled auto-merge (squash) November 14, 2025 17:27
@dcherian dcherian merged commit 33ce95e into pydata:main Nov 14, 2025
36 checks passed
@dcherian dcherian deleted the optimize-coarsen-pad branch November 14, 2025 17:47
shoyer pushed a commit to shoyer/xarray that referenced this pull request Nov 19, 2025
shoyer added a commit that referenced this pull request Nov 20, 2025
* support combine_nested on DataTree objects

* make mypy happy

* Rewrite test_iso8601_decode to work without cftime (#10914)

* Rewrite test_iso8601_decode to work without cftime

Closes #10907

Co-Authored-By: Claude <[email protected]>

* silence warning

---------

Co-authored-by: Claude <[email protected]>

* DOC: Correct minor grammar issues (#10915)

* skip ci based on a label (#10918)

* skip ci based on a label

* retrigger ci

* Optimize padding for coarsening. (#10921)

Co-authored-by: Claude <[email protected]>

* release v2025.11.0 (#10917)

* choose the right version number

* formatting

* move a breaking change entry to the right section

* remove the deprecations section

* missing entry

* remove the empty internals section

* another missing entry

* more missing entries

* remove empty line

* bad merge

* remove the empty documentation section

* newline

* use the old style of setting release dates

* set the release date

* formatting

* back to bold

* release summary

* new release section (#10926)

* new release section

* [skip-ci]

* update whats-new

* removed redundant whatsnew

---------

Co-authored-by: Deepak Cherian <[email protected]>
Co-authored-by: Claude <[email protected]>
Co-authored-by: Christine P. Chai <[email protected]>
Co-authored-by: Justus Magin <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

run-benchmark Run the ASV benchmark workflow topic-performance

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants