perf: cache lexicographic chunk coords in sharding codec
#4012Comparing
d-v-b:perf-sharding-coord-cache
(2742897
) withmain
(6ce787d
)+40%
- Improvement
- 3
- Regression
- 3
- Untouched
- 53
- New
- 10
- Skipped
- 13
No successful run was found on main ( fe22910) during the generation of this report, so 6ce787d was used instead as the comparison base. There might be some changes unrelated to this pull request in this report.
Benchmarks
82 total
test_sharded_morton_single_chunk[(33, 33, 33)-memory]
tests/benchmarks/test_indexing.py-72%
2.2 ms7.7 ms
test_sharded_morton_single_chunk[(32, 32, 32)-memory]
tests/benchmarks/test_indexing.py-69%
2.1 ms7 ms
test_sharded_morton_single_chunk[(30, 30, 30)-memory]
tests/benchmarks/test_indexing.py-66%
2.1 ms6.2 ms
test_sharded_morton_write_single_chunk[(32, 32, 32)-memory]
tests/benchmarks/test_indexing.py×6.5
350.3 ms54 ms
test_sharded_morton_write_single_chunk[(33, 33, 33)-memory]
tests/benchmarks/test_indexing.py×6.3
391.3 ms61.7 ms
test_sharded_morton_write_single_chunk[(30, 30, 30)-memory]
tests/benchmarks/test_indexing.py×6.3
292.7 ms46.7 ms
test_morton_order[(8, 8, 8)]
tests/benchmarks/test_indexing.pyN/A
N/A469.3 µs
test_morton_order[(10, 10, 10)]
tests/benchmarks/test_indexing.pyN/A
N/A861.5 µs
test_morton_order[(20, 20, 20)]
tests/benchmarks/test_indexing.pyN/A
N/A5.6 ms
test_morton_order[(30, 30, 30)]
tests/benchmarks/test_indexing.pyN/A
N/A18.3 ms
test_morton_order[(16, 16, 16)]
tests/benchmarks/test_indexing.pyN/A
N/A2.5 ms
test_morton_order[(32, 32, 32)]
tests/benchmarks/test_indexing.pyN/A
N/A20.8 ms
test_sharded_morton_write_single_chunk_warm_cache[(33, 33, 33)-memory]
tests/benchmarks/test_indexing.pyN/A
N/A37.4 ms
test_morton_order[(33, 33, 33)]
tests/benchmarks/test_indexing.pyN/A
N/A24.6 ms
test_sharded_morton_write_single_chunk_warm_cache[(30, 30, 30)-memory]
tests/benchmarks/test_indexing.pyN/A
N/A28.6 ms
test_sharded_morton_write_single_chunk_warm_cache[(32, 32, 32)-memory]
tests/benchmarks/test_indexing.pyN/A
N/A33.1 ms
test_write_array[memory-Layout(shape=(1000000,), chunks=(1000,), shards=(1000,))-None]
tests/benchmarks/test_e2e.py+2%
1.2 s1.2 s
test_write_array[memory-Layout(shape=(1000000,), chunks=(1000,), shards=(1000,))-gzip]
tests/benchmarks/test_e2e.py+2%
1.8 s1.8 s
test_read_array[local-Layout(shape=(1000000,), chunks=(1000,), shards=None)-gzip]
tests/benchmarks/test_e2e.py+2%
938.1 ms921.4 ms
test_write_array[local-Layout(shape=(1000000,), chunks=(1000,), shards=(1000,))-None]
tests/benchmarks/test_e2e.py+1%
1.9 s1.9 s
test_write_array[memory-Layout(shape=(1000000,), chunks=(1000,), shards=None)-gzip]
tests/benchmarks/test_e2e.py+1%
1.1 s1.1 s
test_write_array[local-Layout(shape=(1000000,), chunks=(100,), shards=(1000000,))-None]
tests/benchmarks/test_e2e.py+1%
5.2 s5.1 s
test_slice_indexing[None-(slice(None, None, None), slice(None, None, None), slice(None, None, None))-memory]
tests/benchmarks/test_indexing.py+1%
354.5 ms350.4 ms
test_slice_indexing[None-(slice(0, None, 4), slice(0, None, 4), slice(0, None, 4))-memory]
tests/benchmarks/test_indexing.py+1%
350.7 ms347 ms
test_write_array[local-Layout(shape=(1000000,), chunks=(1000,), shards=(1000,))-gzip]
tests/benchmarks/test_e2e.py+1%
2.4 s2.4 s
Commits
Click on a commit to change the comparison rangeBase
main
6ce787d
+76.71%
perf: cache lexicographic chunk coords in sharding codec
3372b05
8 days ago
by d-v-b-4.94%
refactor(indexing): make morton_order_iter the lazy primitive too
6840249
1 day ago
by d-v-b-31.45%
perf: build order-coord tuples via .tolist(); document dual representation
2742897
13 hours ago
by d-v-b