zarr-developers/zarr-python - CodSpeed

zarr-developers

zarr-python

Blog Docs Changelog

Performance History

Latest Results

fix/nested shard reads (#3655) * fix partial nested shard reads * changelog

main

3 days ago

remove extra codec pipeline

d-v-b:perf/faster-codecs

3 days ago

Merge branch 'main' into perf/faster-codecs

d-v-b:perf/faster-codecs

4 days ago

fix/nested shard reads (#3655) * fix partial nested shard reads * changelog

main

4 days ago

perf: Optimize Morton order with hypercube and vectorization (#3708) * perf: Skip bounds check for initial elements in 2^n hypercube * lint:Use a list comprehension rather than a for loop * pref:Add decode_morton_vectorized * perf:Replace math.log2() with bit_length() * perf:Use magic numbers for 2D and 3D * perf:Add 4D Morton magic numbers * perf:Add Morton magic numbers for 5D * perf:Remove singleton dimensions to reduce ndims * Add changes * fix:Address type annotation and linting issues * perf:Remove magic number functions * test:Add power of 2 sharding indexing tests * test: Add Morton order benchmarks with cache clearing Add benchmarks that clear the _morton_order LRU cache before each iteration to measure the full Morton computation cost: - test_sharded_morton_indexing: 512-4096 chunks per shard - test_sharded_morton_indexing_large: 32768 chunks per shard Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com> * fix:Bound LRU cache of _morton_order to 16 * test:Add a single chunk test for a large shard * test:Add indexing benchmarks for writing * tests:Add single chunk write test for sharding * Update 3708.misc.md --------- Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com> Co-authored-by: Davis Bennett <davis.v.bennett@gmail.com>

main

4 days ago

Merge branch 'main' into perf/faster-codecs

d-v-b:perf/faster-codecs

4 days ago

default to 1 itemsize for data types that don't declare it

d-v-b:perf/faster-codecs

4 days ago

dynamic pool allocation

d-v-b:perf/faster-codecs

4 days ago

Active Branches

sketch out sync codecs + threadpool

last run

3 days ago

×5

Coalesce and parallelize partial shard reads

last run

7 days ago

+16%

default buffer logic

last run

28 days ago

0%

© 2026 CodSpeed Technology

Home Terms Privacy Docs