Skip to content

Implement fold for iterators for better performance#9412

Draft
Dandandan wants to merge 9 commits intoapache:mainfrom
Dandandan:implement_fold
Draft

Implement fold for iterators for better performance#9412
Dandandan wants to merge 9 commits intoapache:mainfrom
Dandandan:implement_fold

Conversation

@Dandandan
Copy link
Contributor

@Dandandan Dandandan commented Feb 14, 2026

Which issue does this PR close?

Rationale for this change

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

@github-actions github-actions bot added the arrow Changes to the arrow crate label Feb 14, 2026
@Dandandan
Copy link
Contributor Author

run benchmark filter_kernels

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (e8626ef) to d6168e5 diff
BENCH_NAME=filter_kernels
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench filter_kernels
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@Dandandan
Copy link
Contributor Author

run benchmark arrow_reader_clickbench

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                                                         implement_fold                         main
-----                                                                         --------------                         ----
filter context decimal128 (kept 1/2)                                          1.00     43.5±1.48µs        ? ?/sec    1.06     46.2±5.44µs        ? ?/sec
filter context decimal128 high selectivity (kept 1023/1024)                   1.00     49.1±1.19µs        ? ?/sec    1.04     51.2±2.22µs        ? ?/sec
filter context decimal128 low selectivity (kept 1/1024)                       1.00    230.8±0.56ns        ? ?/sec    1.00    231.6±9.85ns        ? ?/sec
filter context f32 (kept 1/2)                                                 1.13     87.6±0.73µs        ? ?/sec    1.00     77.8±0.24µs        ? ?/sec
filter context f32 high selectivity (kept 1023/1024)                          1.00      9.9±0.38µs        ? ?/sec    1.00      9.9±0.38µs        ? ?/sec
filter context f32 low selectivity (kept 1/1024)                              1.01   448.3±26.91ns        ? ?/sec    1.00   444.1±11.55ns        ? ?/sec
filter context fsb with value length 20 (kept 1/2)                            1.17     70.7±0.53µs        ? ?/sec    1.00     60.6±0.17µs        ? ?/sec
filter context fsb with value length 20 high selectivity (kept 1023/1024)     1.17     70.7±1.29µs        ? ?/sec    1.00     60.5±0.08µs        ? ?/sec
filter context fsb with value length 20 low selectivity (kept 1/1024)         1.17     70.6±0.14µs        ? ?/sec    1.00     60.6±0.15µs        ? ?/sec
filter context fsb with value length 5 (kept 1/2)                             1.16     70.6±0.55µs        ? ?/sec    1.00     60.7±0.35µs        ? ?/sec
filter context fsb with value length 5 high selectivity (kept 1023/1024)      1.17     70.8±1.62µs        ? ?/sec    1.00     60.6±0.42µs        ? ?/sec
filter context fsb with value length 5 low selectivity (kept 1/1024)          1.17     70.8±0.86µs        ? ?/sec    1.00     60.5±0.09µs        ? ?/sec
filter context fsb with value length 50 (kept 1/2)                            1.16     70.6±0.13µs        ? ?/sec    1.00     60.6±0.24µs        ? ?/sec
filter context fsb with value length 50 high selectivity (kept 1023/1024)     1.17     70.9±1.19µs        ? ?/sec    1.00     60.6±0.32µs        ? ?/sec
filter context fsb with value length 50 low selectivity (kept 1/1024)         1.19     72.2±7.56µs        ? ?/sec    1.00     60.6±0.17µs        ? ?/sec
filter context i32 (kept 1/2)                                                 1.00     16.5±0.14µs        ? ?/sec    1.01     16.7±0.25µs        ? ?/sec
filter context i32 high selectivity (kept 1023/1024)                          1.02      6.5±0.33µs        ? ?/sec    1.00      6.3±0.30µs        ? ?/sec
filter context i32 low selectivity (kept 1/1024)                              1.01    222.3±1.65ns        ? ?/sec    1.00    220.0±1.27ns        ? ?/sec
filter context i32 w NULLs (kept 1/2)                                         1.12     87.5±0.25µs        ? ?/sec    1.00     77.8±0.43µs        ? ?/sec
filter context i32 w NULLs high selectivity (kept 1023/1024)                  1.05     10.2±0.54µs        ? ?/sec    1.00      9.8±0.31µs        ? ?/sec
filter context i32 w NULLs low selectivity (kept 1/1024)                      1.00    441.6±3.46ns        ? ?/sec    1.00    442.2±1.56ns        ? ?/sec
filter context mixed string view (kept 1/2)                                   1.07    116.2±4.00µs        ? ?/sec    1.00    108.7±3.72µs        ? ?/sec
filter context mixed string view high selectivity (kept 1023/1024)            1.04     55.1±1.74µs        ? ?/sec    1.00     53.2±0.56µs        ? ?/sec
filter context mixed string view low selectivity (kept 1/1024)                1.00    620.3±2.82ns        ? ?/sec    1.00    621.5±2.73ns        ? ?/sec
filter context short string view (kept 1/2)                                   1.10    113.7±4.23µs        ? ?/sec    1.00    103.3±0.83µs        ? ?/sec
filter context short string view high selectivity (kept 1023/1024)            1.04     55.0±1.58µs        ? ?/sec    1.00     52.9±0.99µs        ? ?/sec
filter context short string view low selectivity (kept 1/1024)                1.00    491.9±3.01ns        ? ?/sec    1.00    494.1±2.51ns        ? ?/sec
filter context string (kept 1/2)                                              1.02   583.4±16.12µs        ? ?/sec    1.00   570.7±10.31µs        ? ?/sec
filter context string dictionary (kept 1/2)                                   1.00     17.1±0.05µs        ? ?/sec    1.00     17.1±0.28µs        ? ?/sec
filter context string dictionary high selectivity (kept 1023/1024)            1.00      7.0±0.30µs        ? ?/sec    1.00      7.0±0.25µs        ? ?/sec
filter context string dictionary low selectivity (kept 1/1024)                1.00    641.7±3.40ns        ? ?/sec    1.01   651.0±11.22ns        ? ?/sec
filter context string dictionary w NULLs (kept 1/2)                           1.14     89.0±4.97µs        ? ?/sec    1.00     78.2±0.68µs        ? ?/sec
filter context string dictionary w NULLs high selectivity (kept 1023/1024)    1.01     10.4±0.34µs        ? ?/sec    1.00     10.3±0.35µs        ? ?/sec
filter context string dictionary w NULLs low selectivity (kept 1/1024)        1.00    864.0±4.13ns        ? ?/sec    1.01    873.3±5.82ns        ? ?/sec
filter context string high selectivity (kept 1023/1024)                       1.00   632.6±15.39µs        ? ?/sec    1.06   673.0±13.00µs        ? ?/sec
filter context string low selectivity (kept 1/1024)                           1.07    949.4±3.52ns        ? ?/sec    1.00    889.0±3.52ns        ? ?/sec
filter context u8 (kept 1/2)                                                  1.00     15.0±0.07µs        ? ?/sec    1.00     15.0±0.07µs        ? ?/sec
filter context u8 high selectivity (kept 1023/1024)                           1.11  1988.4±20.27ns        ? ?/sec    1.00  1798.0±10.08ns        ? ?/sec
filter context u8 low selectivity (kept 1/1024)                               1.00    217.1±1.08ns        ? ?/sec    1.00    216.3±0.73ns        ? ?/sec
filter context u8 w NULLs (kept 1/2)                                          1.13     85.8±0.18µs        ? ?/sec    1.00     75.9±0.28µs        ? ?/sec
filter context u8 w NULLs high selectivity (kept 1023/1024)                   1.05      5.4±0.02µs        ? ?/sec    1.00      5.1±0.06µs        ? ?/sec
filter context u8 w NULLs low selectivity (kept 1/1024)                       1.00    437.9±3.19ns        ? ?/sec    1.00    438.9±3.75ns        ? ?/sec
filter decimal128 (kept 1/2)                                                  1.01     61.0±0.46µs        ? ?/sec    1.00     60.2±2.91µs        ? ?/sec
filter decimal128 high selectivity (kept 1023/1024)                           1.00     52.1±0.46µs        ? ?/sec    1.06     55.4±1.64µs        ? ?/sec
filter decimal128 low selectivity (kept 1/1024)                               1.00      2.1±0.01µs        ? ?/sec    1.41      3.0±0.01µs        ? ?/sec
filter f32 (kept 1/2)                                                         1.00    139.1±1.49µs        ? ?/sec    1.18    164.1±0.27µs        ? ?/sec
filter fsb with value length 20 (kept 1/2)                                    1.00    136.4±0.91µs        ? ?/sec    1.05    143.6±0.42µs        ? ?/sec
filter fsb with value length 20 high selectivity (kept 1023/1024)             1.00     69.0±1.21µs        ? ?/sec    1.00     69.3±1.38µs        ? ?/sec
filter fsb with value length 20 low selectivity (kept 1/1024)                 1.48      3.9±0.02µs        ? ?/sec    1.00      2.7±0.01µs        ? ?/sec
filter fsb with value length 5 (kept 1/2)                                     1.00    140.6±1.54µs        ? ?/sec    1.01    141.7±2.63µs        ? ?/sec
filter fsb with value length 5 high selectivity (kept 1023/1024)              1.03     11.4±0.46µs        ? ?/sec    1.00     11.1±0.64µs        ? ?/sec
filter fsb with value length 5 low selectivity (kept 1/1024)                  1.49      3.8±0.11µs        ? ?/sec    1.00      2.6±0.04µs        ? ?/sec
filter fsb with value length 50 (kept 1/2)                                    1.04   178.3±12.77µs        ? ?/sec    1.00   170.7±11.43µs        ? ?/sec
filter fsb with value length 50 high selectivity (kept 1023/1024)             1.01    208.1±5.97µs        ? ?/sec    1.00   206.7±10.10µs        ? ?/sec
filter fsb with value length 50 low selectivity (kept 1/1024)                 1.48      3.9±0.01µs        ? ?/sec    1.00      2.7±0.02µs        ? ?/sec
filter i32 (kept 1/2)                                                         1.00     36.9±0.33µs        ? ?/sec    1.45     53.6±0.18µs        ? ?/sec
filter i32 high selectivity (kept 1023/1024)                                  1.02      8.8±0.41µs        ? ?/sec    1.00      8.6±0.44µs        ? ?/sec
filter i32 low selectivity (kept 1/1024)                                      1.00  1966.7±12.00ns        ? ?/sec    1.50      2.9±0.03µs        ? ?/sec
filter optimize (kept 1/2)                                                    1.66     59.8±1.08µs        ? ?/sec    1.00     35.9±1.18µs        ? ?/sec
filter optimize high selectivity (kept 1023/1024)                             1.00      2.5±0.01µs        ? ?/sec    1.08      2.7±0.02µs        ? ?/sec
filter optimize low selectivity (kept 1/1024)                                 1.00  1794.9±14.44ns        ? ?/sec    1.41      2.5±0.01µs        ? ?/sec
filter run array (kept 1/2)                                                   1.01    450.7±3.59µs        ? ?/sec    1.00    448.3±1.27µs        ? ?/sec
filter run array high selectivity (kept 1023/1024)                            1.01    483.6±4.99µs        ? ?/sec    1.00    478.7±2.30µs        ? ?/sec
filter run array low selectivity (kept 1/1024)                                1.01    368.5±1.18µs        ? ?/sec    1.00    364.5±2.17µs        ? ?/sec
filter single record batch                                                    1.00     37.1±0.33µs        ? ?/sec    1.42     52.5±0.08µs        ? ?/sec
filter u8 (kept 1/2)                                                          1.25     45.3±0.34µs        ? ?/sec    1.00     36.2±0.29µs        ? ?/sec
filter u8 high selectivity (kept 1023/1024)                                   1.04      3.9±0.01µs        ? ?/sec    1.00      3.8±0.02µs        ? ?/sec
filter u8 low selectivity (kept 1/1024)                                       1.00      2.9±0.01µs        ? ?/sec    1.04      3.0±0.01µs        ? ?/sec

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (5888db0) to d6168e5 diff
BENCH_NAME=arrow_reader_clickbench
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench arrow_reader_clickbench
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@Dandandan
Copy link
Contributor Author

run benchmark filter_kernels

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                             implement_fold                         main
-----                                             --------------                         ----
arrow_reader_clickbench/async/Q1                  1.00      2.3±0.01ms        ? ?/sec    1.02      2.3±0.06ms        ? ?/sec
arrow_reader_clickbench/async/Q10                 1.00     10.7±0.20ms        ? ?/sec    1.01     10.8±0.22ms        ? ?/sec
arrow_reader_clickbench/async/Q11                 1.00     12.5±0.21ms        ? ?/sec    1.00     12.5±0.28ms        ? ?/sec
arrow_reader_clickbench/async/Q12                 1.01     22.6±0.26ms        ? ?/sec    1.00     22.3±0.26ms        ? ?/sec
arrow_reader_clickbench/async/Q13                 1.00     27.9±0.23ms        ? ?/sec    1.00     27.9±0.30ms        ? ?/sec
arrow_reader_clickbench/async/Q14                 1.01     25.2±0.34ms        ? ?/sec    1.00     25.0±0.36ms        ? ?/sec
arrow_reader_clickbench/async/Q19                 1.00      5.3±0.07ms        ? ?/sec    1.01      5.4±0.08ms        ? ?/sec
arrow_reader_clickbench/async/Q20                 1.00    114.1±1.06ms        ? ?/sec    1.13    129.1±1.04ms        ? ?/sec
arrow_reader_clickbench/async/Q21                 1.00    129.5±0.63ms        ? ?/sec    1.26    163.5±0.93ms        ? ?/sec
arrow_reader_clickbench/async/Q22                 1.00   267.3±12.24ms        ? ?/sec    1.06   282.1±42.93ms        ? ?/sec
arrow_reader_clickbench/async/Q23                 1.02    406.2±4.32ms        ? ?/sec    1.00    398.2±2.49ms        ? ?/sec
arrow_reader_clickbench/async/Q24                 1.02     31.1±0.29ms        ? ?/sec    1.00     30.6±0.51ms        ? ?/sec
arrow_reader_clickbench/async/Q27                 1.02     98.4±0.34ms        ? ?/sec    1.00     96.3±0.47ms        ? ?/sec
arrow_reader_clickbench/async/Q28                 1.02     96.3±0.97ms        ? ?/sec    1.00     94.6±1.09ms        ? ?/sec
arrow_reader_clickbench/async/Q30                 1.00     27.5±0.25ms        ? ?/sec    1.00     27.5±0.91ms        ? ?/sec
arrow_reader_clickbench/async/Q36                 1.02    116.7±0.85ms        ? ?/sec    1.00    114.2±1.14ms        ? ?/sec
arrow_reader_clickbench/async/Q37                 1.02     92.3±0.55ms        ? ?/sec    1.00     90.2±0.74ms        ? ?/sec
arrow_reader_clickbench/async/Q38                 1.00     34.9±0.22ms        ? ?/sec    1.00     34.8±0.42ms        ? ?/sec
arrow_reader_clickbench/async/Q39                 1.02     45.7±0.30ms        ? ?/sec    1.00     44.8±0.57ms        ? ?/sec
arrow_reader_clickbench/async/Q40                 1.02     40.0±0.85ms        ? ?/sec    1.00     39.3±0.26ms        ? ?/sec
arrow_reader_clickbench/async/Q41                 1.00     28.8±0.57ms        ? ?/sec    1.00     28.9±0.27ms        ? ?/sec
arrow_reader_clickbench/async/Q42                 1.00     10.9±0.20ms        ? ?/sec    1.00     10.9±0.11ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q1     1.00      2.2±0.03ms        ? ?/sec    1.00      2.2±0.01ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q10    1.00     10.3±0.23ms        ? ?/sec    1.01     10.4±0.18ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q11    1.00     11.9±0.25ms        ? ?/sec    1.01     12.0±0.26ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q12    1.03     22.0±0.31ms        ? ?/sec    1.00     21.4±0.21ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q13    1.03     27.0±0.35ms        ? ?/sec    1.00     26.3±0.29ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q14    1.03     24.4±0.29ms        ? ?/sec    1.00     23.8±0.24ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q19    1.01      5.0±0.08ms        ? ?/sec    1.00      5.0±0.07ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q20    1.03    110.3±0.90ms        ? ?/sec    1.00    107.6±1.21ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q21    1.02    125.4±0.84ms        ? ?/sec    1.00    123.0±1.45ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q22    1.03    229.6±0.97ms        ? ?/sec    1.00    223.7±1.29ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q23    1.05    366.0±2.37ms        ? ?/sec    1.00    347.9±2.27ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q24    1.02     29.8±0.41ms        ? ?/sec    1.00     29.2±0.31ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q27    1.02     95.1±1.11ms        ? ?/sec    1.00     93.0±0.86ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q28    1.02     92.6±0.44ms        ? ?/sec    1.00     91.0±0.50ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q30    1.00     26.6±0.31ms        ? ?/sec    1.00     26.5±0.88ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q36    1.02    112.1±0.62ms        ? ?/sec    1.00    110.2±0.55ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q37    1.01     88.8±0.50ms        ? ?/sec    1.00     87.6±0.63ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q38    1.01     32.0±0.32ms        ? ?/sec    1.00     31.7±0.26ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q39    1.01     40.7±0.42ms        ? ?/sec    1.00     40.1±0.45ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q40    1.01     38.3±0.34ms        ? ?/sec    1.00     38.1±0.86ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q41    1.00     27.6±0.36ms        ? ?/sec    1.00     27.5±0.29ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q42    1.00     10.3±0.22ms        ? ?/sec    1.00     10.3±0.12ms        ? ?/sec
arrow_reader_clickbench/sync/Q1                   1.00  1981.8±56.03µs        ? ?/sec    1.01  1993.0±27.46µs        ? ?/sec
arrow_reader_clickbench/sync/Q10                  1.00      7.5±0.07ms        ? ?/sec    1.00      7.5±0.07ms        ? ?/sec
arrow_reader_clickbench/sync/Q11                  1.00      8.9±0.11ms        ? ?/sec    1.00      9.0±0.07ms        ? ?/sec
arrow_reader_clickbench/sync/Q12                  1.00     28.7±0.31ms        ? ?/sec    1.01     29.1±1.31ms        ? ?/sec
arrow_reader_clickbench/sync/Q13                  1.10     36.6±3.69ms        ? ?/sec    1.00     33.4±0.38ms        ? ?/sec
arrow_reader_clickbench/sync/Q14                  1.00     31.2±0.37ms        ? ?/sec    1.23     38.4±0.96ms        ? ?/sec
arrow_reader_clickbench/sync/Q19                  1.01      4.2±0.05ms        ? ?/sec    1.00      4.2±0.04ms        ? ?/sec
arrow_reader_clickbench/sync/Q20                  1.02    177.2±2.56ms        ? ?/sec    1.00    174.2±1.67ms        ? ?/sec
arrow_reader_clickbench/sync/Q21                  1.04    234.1±1.22ms        ? ?/sec    1.00    224.4±4.95ms        ? ?/sec
arrow_reader_clickbench/sync/Q22                  1.02    475.9±3.34ms        ? ?/sec    1.00    468.2±3.36ms        ? ?/sec
arrow_reader_clickbench/sync/Q23                  1.00   440.6±16.72ms        ? ?/sec    1.00   439.0±16.43ms        ? ?/sec
arrow_reader_clickbench/sync/Q24                  1.03     40.3±0.81ms        ? ?/sec    1.00     39.0±0.54ms        ? ?/sec
arrow_reader_clickbench/sync/Q27                  1.03    154.5±1.98ms        ? ?/sec    1.00    149.9±1.54ms        ? ?/sec
arrow_reader_clickbench/sync/Q28                  1.03    147.1±0.94ms        ? ?/sec    1.00    143.5±0.62ms        ? ?/sec
arrow_reader_clickbench/sync/Q30                  1.00     27.6±0.31ms        ? ?/sec    1.00     27.5±0.32ms        ? ?/sec
arrow_reader_clickbench/sync/Q36                  1.02    151.1±1.64ms        ? ?/sec    1.00    148.2±0.85ms        ? ?/sec
arrow_reader_clickbench/sync/Q37                  1.01     85.2±0.68ms        ? ?/sec    1.00     84.0±1.16ms        ? ?/sec
arrow_reader_clickbench/sync/Q38                  1.00     28.2±0.24ms        ? ?/sec    1.00     28.2±0.52ms        ? ?/sec
arrow_reader_clickbench/sync/Q39                  1.02     33.3±0.58ms        ? ?/sec    1.00     32.8±0.21ms        ? ?/sec
arrow_reader_clickbench/sync/Q40                  1.03     26.2±0.37ms        ? ?/sec    1.00     25.5±0.26ms        ? ?/sec
arrow_reader_clickbench/sync/Q41                  1.00     27.4±0.37ms        ? ?/sec    1.00     27.3±0.59ms        ? ?/sec
arrow_reader_clickbench/sync/Q42                  1.00     11.6±0.11ms        ? ?/sec    1.01     11.7±0.27ms        ? ?/sec

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (420d8ba) to d6168e5 diff
BENCH_NAME=filter_kernels
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench filter_kernels
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                                                         implement_fold                         main
-----                                                                         --------------                         ----
filter context decimal128 (kept 1/2)                                          1.00     43.4±4.55µs        ? ?/sec    1.21     52.5±7.84µs        ? ?/sec
filter context decimal128 high selectivity (kept 1023/1024)                   1.04     51.4±1.73µs        ? ?/sec    1.00     49.5±1.51µs        ? ?/sec
filter context decimal128 low selectivity (kept 1/1024)                       1.00    234.8±2.54ns        ? ?/sec    1.06    248.5±4.86ns        ? ?/sec
filter context f32 (kept 1/2)                                                 1.12     87.4±0.36µs        ? ?/sec    1.00     77.7±0.21µs        ? ?/sec
filter context f32 high selectivity (kept 1023/1024)                          1.02     10.4±0.46µs        ? ?/sec    1.00     10.2±0.52µs        ? ?/sec
filter context f32 low selectivity (kept 1/1024)                              1.00    423.7±4.15ns        ? ?/sec    1.10    466.1±7.88ns        ? ?/sec
filter context fsb with value length 20 (kept 1/2)                            1.17     70.9±2.14µs        ? ?/sec    1.00     60.6±0.25µs        ? ?/sec
filter context fsb with value length 20 high selectivity (kept 1023/1024)     1.17     70.6±0.68µs        ? ?/sec    1.00     60.6±0.20µs        ? ?/sec
filter context fsb with value length 20 low selectivity (kept 1/1024)         1.17     70.6±0.36µs        ? ?/sec    1.00     60.5±0.15µs        ? ?/sec
filter context fsb with value length 5 (kept 1/2)                             1.16     70.6±0.08µs        ? ?/sec    1.00     60.8±1.22µs        ? ?/sec
filter context fsb with value length 5 high selectivity (kept 1023/1024)      1.16     70.7±0.66µs        ? ?/sec    1.00     60.7±1.12µs        ? ?/sec
filter context fsb with value length 5 low selectivity (kept 1/1024)          1.16     70.5±0.14µs        ? ?/sec    1.00     60.6±0.20µs        ? ?/sec
filter context fsb with value length 50 (kept 1/2)                            1.17     70.9±1.45µs        ? ?/sec    1.00     60.7±0.11µs        ? ?/sec
filter context fsb with value length 50 high selectivity (kept 1023/1024)     1.16     70.7±0.94µs        ? ?/sec    1.00     61.0±1.46µs        ? ?/sec
filter context fsb with value length 50 low selectivity (kept 1/1024)         1.16     70.6±0.21µs        ? ?/sec    1.00     60.7±0.97µs        ? ?/sec
filter context i32 (kept 1/2)                                                 1.00     16.3±0.08µs        ? ?/sec    1.01     16.4±0.13µs        ? ?/sec
filter context i32 high selectivity (kept 1023/1024)                          1.00      6.4±0.43µs        ? ?/sec    1.05      6.7±0.27µs        ? ?/sec
filter context i32 low selectivity (kept 1/1024)                              1.00    225.1±4.03ns        ? ?/sec    1.06    237.6±4.91ns        ? ?/sec
filter context i32 w NULLs (kept 1/2)                                         1.13     87.8±0.92µs        ? ?/sec    1.00     77.8±0.20µs        ? ?/sec
filter context i32 w NULLs high selectivity (kept 1023/1024)                  1.00      9.8±0.42µs        ? ?/sec    1.06     10.4±0.37µs        ? ?/sec
filter context i32 w NULLs low selectivity (kept 1/1024)                      1.00    423.9±3.88ns        ? ?/sec    1.10    465.7±5.48ns        ? ?/sec
filter context mixed string view (kept 1/2)                                   1.10    112.6±2.53µs        ? ?/sec    1.00    102.2±2.16µs        ? ?/sec
filter context mixed string view high selectivity (kept 1023/1024)            1.00     51.7±0.55µs        ? ?/sec    1.03     53.3±0.68µs        ? ?/sec
filter context mixed string view low selectivity (kept 1/1024)                1.00    614.4±7.52ns        ? ?/sec    1.07    658.4±3.87ns        ? ?/sec
filter context short string view (kept 1/2)                                   1.11    115.5±5.05µs        ? ?/sec    1.00    103.6±4.78µs        ? ?/sec
filter context short string view high selectivity (kept 1023/1024)            1.01     53.1±0.66µs        ? ?/sec    1.00     52.5±1.05µs        ? ?/sec
filter context short string view low selectivity (kept 1/1024)                1.00    479.3±1.57ns        ? ?/sec    1.09    520.4±5.45ns        ? ?/sec
filter context string (kept 1/2)                                              1.00   584.2±14.33µs        ? ?/sec    1.00   582.4±13.86µs        ? ?/sec
filter context string dictionary (kept 1/2)                                   1.01     17.0±0.14µs        ? ?/sec    1.00     16.9±0.05µs        ? ?/sec
filter context string dictionary high selectivity (kept 1023/1024)            1.09      7.6±0.27µs        ? ?/sec    1.00      6.9±0.26µs        ? ?/sec
filter context string dictionary low selectivity (kept 1/1024)                1.00    639.7±4.74ns        ? ?/sec    1.03    660.2±4.61ns        ? ?/sec
filter context string dictionary w NULLs (kept 1/2)                           1.12     87.9±0.37µs        ? ?/sec    1.00     78.4±0.46µs        ? ?/sec
filter context string dictionary w NULLs high selectivity (kept 1023/1024)    1.00     10.4±0.42µs        ? ?/sec    1.00     10.4±0.48µs        ? ?/sec
filter context string dictionary w NULLs low selectivity (kept 1/1024)        1.00   863.4±11.43ns        ? ?/sec    1.03    890.6±6.76ns        ? ?/sec
filter context string high selectivity (kept 1023/1024)                       1.00   641.5±11.97µs        ? ?/sec    1.07   685.0±17.47µs        ? ?/sec
filter context string low selectivity (kept 1/1024)                           1.00    898.3±3.65ns        ? ?/sec    1.03    926.8±4.04ns        ? ?/sec
filter context u8 (kept 1/2)                                                  1.00     15.0±0.08µs        ? ?/sec    1.00     15.0±0.03µs        ? ?/sec
filter context u8 high selectivity (kept 1023/1024)                           1.03  1993.4±10.45ns        ? ?/sec    1.00  1936.1±13.07ns        ? ?/sec
filter context u8 low selectivity (kept 1/1024)                               1.00    214.3±1.33ns        ? ?/sec    1.11    237.2±4.71ns        ? ?/sec
filter context u8 w NULLs (kept 1/2)                                          1.13     85.9±0.66µs        ? ?/sec    1.00     76.0±0.72µs        ? ?/sec
filter context u8 w NULLs high selectivity (kept 1023/1024)                   1.01      5.2±0.03µs        ? ?/sec    1.00      5.1±0.02µs        ? ?/sec
filter context u8 w NULLs low selectivity (kept 1/1024)                       1.00    418.9±1.79ns        ? ?/sec    1.12    468.1±4.88ns        ? ?/sec
filter decimal128 (kept 1/2)                                                  1.00     42.0±3.72µs        ? ?/sec    1.38     58.0±3.38µs        ? ?/sec
filter decimal128 high selectivity (kept 1023/1024)                           1.00     50.8±0.74µs        ? ?/sec    1.06     54.0±1.59µs        ? ?/sec
filter decimal128 low selectivity (kept 1/1024)                               1.00   1850.6±4.97ns        ? ?/sec    1.61      3.0±0.05µs        ? ?/sec
filter f32 (kept 1/2)                                                         1.00    133.8±0.39µs        ? ?/sec    1.23    164.3±0.48µs        ? ?/sec
filter fsb with value length 20 (kept 1/2)                                    1.00    126.0±0.20µs        ? ?/sec    1.14    144.0±1.09µs        ? ?/sec
filter fsb with value length 20 high selectivity (kept 1023/1024)             1.00     68.5±2.29µs        ? ?/sec    1.05     71.7±2.15µs        ? ?/sec
filter fsb with value length 20 low selectivity (kept 1/1024)                 1.00      2.2±0.07µs        ? ?/sec    1.22      2.7±0.04µs        ? ?/sec
filter fsb with value length 5 (kept 1/2)                                     1.00    132.8±0.24µs        ? ?/sec    1.07    142.1±0.59µs        ? ?/sec
filter fsb with value length 5 high selectivity (kept 1023/1024)              1.00     10.3±0.65µs        ? ?/sec    1.06     10.9±0.49µs        ? ?/sec
filter fsb with value length 5 low selectivity (kept 1/1024)                  1.00      2.2±0.01µs        ? ?/sec    1.19      2.6±0.03µs        ? ?/sec
filter fsb with value length 50 (kept 1/2)                                    1.00    147.0±2.58µs        ? ?/sec    1.09    159.6±4.01µs        ? ?/sec
filter fsb with value length 50 high selectivity (kept 1023/1024)             1.00    210.5±7.24µs        ? ?/sec    1.06    223.9±6.84µs        ? ?/sec
filter fsb with value length 50 low selectivity (kept 1/1024)                 1.00      2.2±0.01µs        ? ?/sec    1.21      2.6±0.01µs        ? ?/sec
filter i32 (kept 1/2)                                                         1.00     32.0±0.14µs        ? ?/sec    1.68     53.6±0.14µs        ? ?/sec
filter i32 high selectivity (kept 1023/1024)                                  1.00      7.6±0.44µs        ? ?/sec    1.15      8.7±0.32µs        ? ?/sec
filter i32 low selectivity (kept 1/1024)                                      1.00   1897.6±7.87ns        ? ?/sec    1.55      2.9±0.03µs        ? ?/sec
filter optimize (kept 1/2)                                                    1.11     39.7±0.06µs        ? ?/sec    1.00     35.8±0.16µs        ? ?/sec
filter optimize high selectivity (kept 1023/1024)                             1.00      2.5±0.01µs        ? ?/sec    1.09      2.8±0.03µs        ? ?/sec
filter optimize low selectivity (kept 1/1024)                                 1.00  1730.0±36.44ns        ? ?/sec    1.46      2.5±0.02µs        ? ?/sec
filter run array (kept 1/2)                                                   1.00    442.5±2.68µs        ? ?/sec    1.01    446.6±2.00µs        ? ?/sec
filter run array high selectivity (kept 1023/1024)                            1.00    477.3±4.05µs        ? ?/sec    1.01   481.6±14.12µs        ? ?/sec
filter run array low selectivity (kept 1/1024)                                1.00    364.2±4.63µs        ? ?/sec    1.00    364.3±2.39µs        ? ?/sec
filter single record batch                                                    1.00     32.0±0.19µs        ? ?/sec    1.65     52.8±1.04µs        ? ?/sec
filter u8 (kept 1/2)                                                          1.00     34.1±0.07µs        ? ?/sec    1.06     36.2±0.57µs        ? ?/sec
filter u8 high selectivity (kept 1023/1024)                                   1.00      3.2±0.09µs        ? ?/sec    1.17      3.7±0.07µs        ? ?/sec
filter u8 low selectivity (kept 1/1024)                                       1.00      2.3±0.09µs        ? ?/sec    1.31      3.0±0.01µs        ? ?/sec

@Dandandan
Copy link
Contributor Author

run benchmark arrow_reader_clickbench

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (420d8ba) to d6168e5 diff
BENCH_NAME=arrow_reader_clickbench
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench arrow_reader_clickbench
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                             implement_fold                         main
-----                                             --------------                         ----
arrow_reader_clickbench/async/Q1                  1.00      2.3±0.06ms        ? ?/sec    1.01      2.4±0.03ms        ? ?/sec
arrow_reader_clickbench/async/Q10                 1.00     10.7±0.26ms        ? ?/sec    1.01     10.8±0.22ms        ? ?/sec
arrow_reader_clickbench/async/Q11                 1.00     12.3±0.25ms        ? ?/sec    1.01     12.5±0.48ms        ? ?/sec
arrow_reader_clickbench/async/Q12                 1.00     22.4±0.43ms        ? ?/sec    1.01     22.6±0.32ms        ? ?/sec
arrow_reader_clickbench/async/Q13                 1.00     27.9±0.28ms        ? ?/sec    1.01     28.3±0.81ms        ? ?/sec
arrow_reader_clickbench/async/Q14                 1.00     24.9±0.28ms        ? ?/sec    1.01     25.0±0.52ms        ? ?/sec
arrow_reader_clickbench/async/Q19                 1.00      5.4±0.08ms        ? ?/sec    1.02      5.5±0.19ms        ? ?/sec
arrow_reader_clickbench/async/Q20                 1.23    155.7±1.15ms        ? ?/sec    1.00    126.9±9.74ms        ? ?/sec
arrow_reader_clickbench/async/Q21                 1.00    162.9±1.02ms        ? ?/sec    1.00    162.8±1.10ms        ? ?/sec
arrow_reader_clickbench/async/Q22                 1.02    238.5±1.17ms        ? ?/sec    1.00    234.7±1.49ms        ? ?/sec
arrow_reader_clickbench/async/Q23                 1.00    404.8±2.36ms        ? ?/sec    1.00    403.3±2.89ms        ? ?/sec
arrow_reader_clickbench/async/Q24                 1.00     31.2±0.38ms        ? ?/sec    1.00     31.2±0.33ms        ? ?/sec
arrow_reader_clickbench/async/Q27                 1.02     99.2±0.94ms        ? ?/sec    1.00     97.5±0.40ms        ? ?/sec
arrow_reader_clickbench/async/Q28                 1.00     96.3±0.47ms        ? ?/sec    1.00     96.1±0.59ms        ? ?/sec
arrow_reader_clickbench/async/Q30                 1.00     27.3±0.32ms        ? ?/sec    1.01     27.6±0.72ms        ? ?/sec
arrow_reader_clickbench/async/Q36                 1.01    116.7±0.66ms        ? ?/sec    1.00    116.0±0.59ms        ? ?/sec
arrow_reader_clickbench/async/Q37                 1.00     92.6±0.74ms        ? ?/sec    1.00     92.2±0.72ms        ? ?/sec
arrow_reader_clickbench/async/Q38                 1.01     35.2±0.55ms        ? ?/sec    1.00     34.8±0.42ms        ? ?/sec
arrow_reader_clickbench/async/Q39                 1.02     45.7±0.30ms        ? ?/sec    1.00     45.0±0.53ms        ? ?/sec
arrow_reader_clickbench/async/Q40                 1.01     39.5±0.49ms        ? ?/sec    1.00     39.2±0.39ms        ? ?/sec
arrow_reader_clickbench/async/Q41                 1.00     28.7±0.41ms        ? ?/sec    1.00     28.6±0.32ms        ? ?/sec
arrow_reader_clickbench/async/Q42                 1.00     10.8±0.10ms        ? ?/sec    1.03     11.1±0.14ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q1     1.00      2.2±0.01ms        ? ?/sec    1.01      2.2±0.01ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q10    1.00     10.3±0.27ms        ? ?/sec    1.02     10.5±0.23ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q11    1.00     11.7±0.18ms        ? ?/sec    1.03     12.1±0.30ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q12    1.00     21.7±0.33ms        ? ?/sec    1.01     21.9±0.23ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q13    1.00     26.9±0.45ms        ? ?/sec    1.00     27.0±0.31ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q14    1.00     24.4±0.27ms        ? ?/sec    1.01     24.7±0.23ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q19    1.00      5.0±0.07ms        ? ?/sec    1.01      5.1±0.06ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q20    1.01    109.5±0.81ms        ? ?/sec    1.00    108.4±0.50ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q21    1.01    124.7±1.11ms        ? ?/sec    1.00    123.9±0.86ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q22    1.01    229.2±1.02ms        ? ?/sec    1.00    227.9±1.71ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q23    1.04    366.0±2.15ms        ? ?/sec    1.00    351.8±2.03ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q24    1.00     29.4±0.69ms        ? ?/sec    1.01     29.7±0.36ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q27    1.01     94.3±0.47ms        ? ?/sec    1.00     93.4±0.55ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q28    1.01     91.9±0.44ms        ? ?/sec    1.00     91.2±0.43ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q30    1.00     25.9±0.24ms        ? ?/sec    1.01     26.3±0.26ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q36    1.01    111.9±1.25ms        ? ?/sec    1.00    110.7±0.55ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q37    1.01     89.3±1.21ms        ? ?/sec    1.00     88.1±1.20ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q38    1.00     31.7±0.29ms        ? ?/sec    1.00     31.6±0.48ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q39    1.03     40.7±0.40ms        ? ?/sec    1.00     39.7±0.27ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q40    1.00     37.7±0.66ms        ? ?/sec    1.00     37.7±0.39ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q41    1.01     27.4±0.31ms        ? ?/sec    1.00     27.2±0.21ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q42    1.01     10.3±0.24ms        ? ?/sec    1.00     10.2±0.08ms        ? ?/sec
arrow_reader_clickbench/sync/Q1                   1.00   1965.2±8.02µs        ? ?/sec    1.01   1987.7±7.50µs        ? ?/sec
arrow_reader_clickbench/sync/Q10                  1.00      7.5±0.26ms        ? ?/sec    1.01      7.5±0.03ms        ? ?/sec
arrow_reader_clickbench/sync/Q11                  1.00      8.9±0.14ms        ? ?/sec    1.01      8.9±0.06ms        ? ?/sec
arrow_reader_clickbench/sync/Q12                  1.01     29.2±0.95ms        ? ?/sec    1.00     28.9±1.32ms        ? ?/sec
arrow_reader_clickbench/sync/Q13                  1.30     43.6±1.04ms        ? ?/sec    1.00     33.5±0.27ms        ? ?/sec
arrow_reader_clickbench/sync/Q14                  1.00     31.1±0.35ms        ? ?/sec    1.24     38.5±1.07ms        ? ?/sec
arrow_reader_clickbench/sync/Q19                  1.02      4.2±0.19ms        ? ?/sec    1.00      4.2±0.02ms        ? ?/sec
arrow_reader_clickbench/sync/Q20                  1.01    176.2±0.86ms        ? ?/sec    1.00    174.5±1.17ms        ? ?/sec
arrow_reader_clickbench/sync/Q21                  1.02    229.8±5.25ms        ? ?/sec    1.00    224.9±5.25ms        ? ?/sec
arrow_reader_clickbench/sync/Q22                  1.01    473.5±2.95ms        ? ?/sec    1.00    470.6±2.96ms        ? ?/sec
arrow_reader_clickbench/sync/Q23                  1.00   441.3±19.01ms        ? ?/sec    1.00   441.0±17.06ms        ? ?/sec
arrow_reader_clickbench/sync/Q24                  1.01     39.4±0.94ms        ? ?/sec    1.00     39.0±0.30ms        ? ?/sec
arrow_reader_clickbench/sync/Q27                  1.03    153.7±0.92ms        ? ?/sec    1.00    148.9±0.61ms        ? ?/sec
arrow_reader_clickbench/sync/Q28                  1.02    147.2±1.29ms        ? ?/sec    1.00    143.7±1.26ms        ? ?/sec
arrow_reader_clickbench/sync/Q30                  1.00     27.8±0.40ms        ? ?/sec    1.00     27.7±0.45ms        ? ?/sec
arrow_reader_clickbench/sync/Q36                  1.02    151.5±1.12ms        ? ?/sec    1.00    148.1±1.33ms        ? ?/sec
arrow_reader_clickbench/sync/Q37                  1.01     85.0±0.66ms        ? ?/sec    1.00     84.0±0.49ms        ? ?/sec
arrow_reader_clickbench/sync/Q38                  1.01     28.1±0.19ms        ? ?/sec    1.00     27.9±0.34ms        ? ?/sec
arrow_reader_clickbench/sync/Q39                  1.03     33.5±0.37ms        ? ?/sec    1.00     32.5±0.47ms        ? ?/sec
arrow_reader_clickbench/sync/Q40                  1.02     26.1±0.32ms        ? ?/sec    1.00     25.6±0.82ms        ? ?/sec
arrow_reader_clickbench/sync/Q41                  1.00     27.2±0.53ms        ? ?/sec    1.00     27.1±0.35ms        ? ?/sec
arrow_reader_clickbench/sync/Q42                  1.00     11.6±0.31ms        ? ?/sec    1.00     11.6±0.09ms        ? ?/sec

@Dandandan
Copy link
Contributor Author

run benchmark filter_kernels

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (9e4c209) to d6168e5 diff
BENCH_NAME=filter_kernels
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench filter_kernels
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                                                         implement_fold                         main
-----                                                                         --------------                         ----
filter context decimal128 (kept 1/2)                                          1.00     43.1±1.06µs        ? ?/sec    1.11     47.9±7.20µs        ? ?/sec
filter context decimal128 high selectivity (kept 1023/1024)                   1.02     49.0±1.07µs        ? ?/sec    1.00     48.2±1.20µs        ? ?/sec
filter context decimal128 low selectivity (kept 1/1024)                       1.02    240.1±5.24ns        ? ?/sec    1.00    234.3±5.10ns        ? ?/sec
filter context f32 (kept 1/2)                                                 1.13     87.8±0.98µs        ? ?/sec    1.00     77.8±0.57µs        ? ?/sec
filter context f32 high selectivity (kept 1023/1024)                          1.07     10.6±0.31µs        ? ?/sec    1.00      9.9±0.33µs        ? ?/sec
filter context f32 low selectivity (kept 1/1024)                              1.00    430.4±6.36ns        ? ?/sec    1.03    444.0±5.09ns        ? ?/sec
filter context fsb with value length 20 (kept 1/2)                            1.17     70.6±0.19µs        ? ?/sec    1.00     60.6±0.10µs        ? ?/sec
filter context fsb with value length 20 high selectivity (kept 1023/1024)     1.18     71.3±4.01µs        ? ?/sec    1.00     60.6±0.17µs        ? ?/sec
filter context fsb with value length 20 low selectivity (kept 1/1024)         1.17     70.6±0.47µs        ? ?/sec    1.00     60.5±0.08µs        ? ?/sec
filter context fsb with value length 5 (kept 1/2)                             1.16     70.6±0.64µs        ? ?/sec    1.00     60.8±0.86µs        ? ?/sec
filter context fsb with value length 5 high selectivity (kept 1023/1024)      1.16     70.6±0.39µs        ? ?/sec    1.00     60.6±0.67µs        ? ?/sec
filter context fsb with value length 5 low selectivity (kept 1/1024)          1.16     70.5±0.17µs        ? ?/sec    1.00     60.6±0.37µs        ? ?/sec
filter context fsb with value length 50 (kept 1/2)                            1.17     70.8±1.45µs        ? ?/sec    1.00     60.7±0.62µs        ? ?/sec
filter context fsb with value length 50 high selectivity (kept 1023/1024)     1.16     70.6±0.31µs        ? ?/sec    1.00     60.7±0.70µs        ? ?/sec
filter context fsb with value length 50 low selectivity (kept 1/1024)         1.16     70.7±0.28µs        ? ?/sec    1.00     60.7±0.55µs        ? ?/sec
filter context i32 (kept 1/2)                                                 1.00     16.3±0.12µs        ? ?/sec    1.01     16.5±0.06µs        ? ?/sec
filter context i32 high selectivity (kept 1023/1024)                          1.00      6.3±0.27µs        ? ?/sec    1.06      6.7±0.46µs        ? ?/sec
filter context i32 low selectivity (kept 1/1024)                              1.02    227.7±5.52ns        ? ?/sec    1.00    224.0±4.74ns        ? ?/sec
filter context i32 w NULLs (kept 1/2)                                         1.13     87.8±1.17µs        ? ?/sec    1.00     77.8±0.95µs        ? ?/sec
filter context i32 w NULLs high selectivity (kept 1023/1024)                  1.04     10.1±0.43µs        ? ?/sec    1.00      9.7±0.42µs        ? ?/sec
filter context i32 w NULLs low selectivity (kept 1/1024)                      1.00    428.6±6.84ns        ? ?/sec    1.03    443.3±4.98ns        ? ?/sec
filter context mixed string view (kept 1/2)                                   1.08    113.5±2.53µs        ? ?/sec    1.00    105.2±5.66µs        ? ?/sec
filter context mixed string view high selectivity (kept 1023/1024)            1.00     54.0±1.76µs        ? ?/sec    1.01     54.3±1.63µs        ? ?/sec
filter context mixed string view low selectivity (kept 1/1024)                1.00    612.8±4.81ns        ? ?/sec    1.02   627.0±10.57ns        ? ?/sec
filter context short string view (kept 1/2)                                   1.08    114.3±3.73µs        ? ?/sec    1.00    105.5±3.98µs        ? ?/sec
filter context short string view high selectivity (kept 1023/1024)            1.00     53.2±1.21µs        ? ?/sec    1.00     53.4±1.57µs        ? ?/sec
filter context short string view low selectivity (kept 1/1024)                1.00    476.3±4.36ns        ? ?/sec    1.04    495.8±7.57ns        ? ?/sec
filter context string (kept 1/2)                                              1.00   579.2±10.23µs        ? ?/sec    1.00   580.1±14.84µs        ? ?/sec
filter context string dictionary (kept 1/2)                                   1.00     16.8±0.15µs        ? ?/sec    1.00     16.8±0.16µs        ? ?/sec
filter context string dictionary high selectivity (kept 1023/1024)            1.02      7.3±0.32µs        ? ?/sec    1.00      7.1±0.27µs        ? ?/sec
filter context string dictionary low selectivity (kept 1/1024)                1.00    636.8±6.01ns        ? ?/sec    1.03    656.5±9.75ns        ? ?/sec
filter context string dictionary w NULLs (kept 1/2)                           1.13     88.2±1.42µs        ? ?/sec    1.00     78.4±1.23µs        ? ?/sec
filter context string dictionary w NULLs high selectivity (kept 1023/1024)    1.02     10.5±0.49µs        ? ?/sec    1.00     10.3±0.31µs        ? ?/sec
filter context string dictionary w NULLs low selectivity (kept 1/1024)        1.00    848.4±9.10ns        ? ?/sec    1.03    876.2±5.26ns        ? ?/sec
filter context string high selectivity (kept 1023/1024)                       1.00   636.2±14.26µs        ? ?/sec    1.06   673.6±12.10µs        ? ?/sec
filter context string low selectivity (kept 1/1024)                           1.02    901.6±7.00ns        ? ?/sec    1.00    884.4±6.99ns        ? ?/sec
filter context u8 (kept 1/2)                                                  1.00     15.0±0.04µs        ? ?/sec    1.00     15.0±0.12µs        ? ?/sec
filter context u8 high selectivity (kept 1023/1024)                           1.12      2.0±0.01µs        ? ?/sec    1.00  1798.6±16.49ns        ? ?/sec
filter context u8 low selectivity (kept 1/1024)                               1.01    221.4±5.81ns        ? ?/sec    1.00    219.9±4.72ns        ? ?/sec
filter context u8 w NULLs (kept 1/2)                                          1.14     86.1±1.17µs        ? ?/sec    1.00     75.8±0.12µs        ? ?/sec
filter context u8 w NULLs high selectivity (kept 1023/1024)                   1.06      5.4±0.02µs        ? ?/sec    1.00      5.1±0.05µs        ? ?/sec
filter context u8 w NULLs low selectivity (kept 1/1024)                       1.00    420.5±5.97ns        ? ?/sec    1.05    441.8±6.72ns        ? ?/sec
filter decimal128 (kept 1/2)                                                  1.62     90.6±0.59µs        ? ?/sec    1.00     55.9±1.33µs        ? ?/sec
filter decimal128 high selectivity (kept 1023/1024)                           1.00     50.4±0.73µs        ? ?/sec    1.01     50.9±1.22µs        ? ?/sec
filter decimal128 low selectivity (kept 1/1024)                               1.00      2.5±0.01µs        ? ?/sec    1.19      3.0±0.01µs        ? ?/sec
filter f32 (kept 1/2)                                                         1.16    190.0±0.53µs        ? ?/sec    1.00    164.2±0.33µs        ? ?/sec
filter fsb with value length 20 (kept 1/2)                                    1.02    146.5±1.86µs        ? ?/sec    1.00    143.7±0.53µs        ? ?/sec
filter fsb with value length 20 high selectivity (kept 1023/1024)             1.00     67.2±0.65µs        ? ?/sec    1.03     69.3±2.39µs        ? ?/sec
filter fsb with value length 20 low selectivity (kept 1/1024)                 1.60      4.3±0.04µs        ? ?/sec    1.00      2.7±0.01µs        ? ?/sec
filter fsb with value length 5 (kept 1/2)                                     1.05    149.1±0.88µs        ? ?/sec    1.00    141.3±0.47µs        ? ?/sec
filter fsb with value length 5 high selectivity (kept 1023/1024)              1.00     10.0±0.59µs        ? ?/sec    1.11     11.1±0.71µs        ? ?/sec
filter fsb with value length 5 low selectivity (kept 1/1024)                  1.59      4.2±0.04µs        ? ?/sec    1.00      2.6±0.01µs        ? ?/sec
filter fsb with value length 50 (kept 1/2)                                    1.11    172.8±5.48µs        ? ?/sec    1.00    155.7±1.69µs        ? ?/sec
filter fsb with value length 50 high selectivity (kept 1023/1024)             1.05    222.8±8.75µs        ? ?/sec    1.00   212.0±12.62µs        ? ?/sec
filter fsb with value length 50 low selectivity (kept 1/1024)                 1.61      4.3±0.05µs        ? ?/sec    1.00      2.7±0.03µs        ? ?/sec
filter i32 (kept 1/2)                                                         1.65     88.4±0.74µs        ? ?/sec    1.00     53.6±0.08µs        ? ?/sec
filter i32 high selectivity (kept 1023/1024)                                  1.00      7.6±0.47µs        ? ?/sec    1.12      8.6±0.30µs        ? ?/sec
filter i32 low selectivity (kept 1/1024)                                      1.00      2.4±0.01µs        ? ?/sec    1.20      2.9±0.03µs        ? ?/sec
filter optimize (kept 1/2)                                                    1.00     34.9±0.27µs        ? ?/sec    1.02     35.7±0.07µs        ? ?/sec
filter optimize high selectivity (kept 1023/1024)                             1.00      2.6±0.02µs        ? ?/sec    1.08      2.8±0.01µs        ? ?/sec
filter optimize low selectivity (kept 1/1024)                                 1.00   1270.7±4.23ns        ? ?/sec    1.99      2.5±0.01µs        ? ?/sec
filter run array (kept 1/2)                                                   1.10    493.9±1.68µs        ? ?/sec    1.00    448.8±6.21µs        ? ?/sec
filter run array high selectivity (kept 1023/1024)                            1.01    481.3±2.09µs        ? ?/sec    1.00    477.8±5.76µs        ? ?/sec
filter run array low selectivity (kept 1/1024)                                1.01    369.2±1.30µs        ? ?/sec    1.00    364.2±3.46µs        ? ?/sec
filter single record batch                                                    1.68     88.3±0.79µs        ? ?/sec    1.00     52.6±0.57µs        ? ?/sec
filter u8 (kept 1/2)                                                          2.22     80.1±0.64µs        ? ?/sec    1.00     36.0±0.06µs        ? ?/sec
filter u8 high selectivity (kept 1023/1024)                                   1.00      3.2±0.02µs        ? ?/sec    1.18      3.7±0.02µs        ? ?/sec
filter u8 low selectivity (kept 1/1024)                                       1.00      2.5±0.02µs        ? ?/sec    1.18      3.0±0.10µs        ? ?/sec

@Dandandan
Copy link
Contributor Author

run benchmark filter_kernels

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (ea57375) to d6168e5 diff
BENCH_NAME=filter_kernels
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench filter_kernels
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                                                         implement_fold                         main
-----                                                                         --------------                         ----
filter context decimal128 (kept 1/2)                                          1.01     45.1±5.71µs        ? ?/sec    1.00     44.8±5.65µs        ? ?/sec
filter context decimal128 high selectivity (kept 1023/1024)                   1.08     52.9±1.64µs        ? ?/sec    1.00     49.2±1.16µs        ? ?/sec
filter context decimal128 low selectivity (kept 1/1024)                       1.00    229.9±2.98ns        ? ?/sec    1.00    229.0±8.83ns        ? ?/sec
filter context f32 (kept 1/2)                                                 1.13     87.4±0.19µs        ? ?/sec    1.00     77.5±0.49µs        ? ?/sec
filter context f32 high selectivity (kept 1023/1024)                          1.04     10.3±0.38µs        ? ?/sec    1.00      9.9±0.36µs        ? ?/sec
filter context f32 low selectivity (kept 1/1024)                              1.00    441.3±1.54ns        ? ?/sec    1.01   446.7±32.56ns        ? ?/sec
filter context fsb with value length 20 (kept 1/2)                            1.17     70.8±0.46µs        ? ?/sec    1.00     60.5±0.23µs        ? ?/sec
filter context fsb with value length 20 high selectivity (kept 1023/1024)     1.17     70.7±0.40µs        ? ?/sec    1.00     60.5±0.21µs        ? ?/sec
filter context fsb with value length 20 low selectivity (kept 1/1024)         1.17     70.7±0.33µs        ? ?/sec    1.00     60.6±0.29µs        ? ?/sec
filter context fsb with value length 5 (kept 1/2)                             1.16     70.7±0.19µs        ? ?/sec    1.00     60.7±0.78µs        ? ?/sec
filter context fsb with value length 5 high selectivity (kept 1023/1024)      1.17     70.7±0.19µs        ? ?/sec    1.00     60.6±0.79µs        ? ?/sec
filter context fsb with value length 5 low selectivity (kept 1/1024)          1.17     70.6±0.22µs        ? ?/sec    1.00     60.5±0.11µs        ? ?/sec
filter context fsb with value length 50 (kept 1/2)                            1.16     70.6±0.40µs        ? ?/sec    1.00     60.6±0.60µs        ? ?/sec
filter context fsb with value length 50 high selectivity (kept 1023/1024)     1.17     70.6±0.39µs        ? ?/sec    1.00     60.6±0.21µs        ? ?/sec
filter context fsb with value length 50 low selectivity (kept 1/1024)         1.17     70.6±0.16µs        ? ?/sec    1.00     60.6±0.18µs        ? ?/sec
filter context i32 (kept 1/2)                                                 1.00     16.4±0.06µs        ? ?/sec    1.02     16.7±0.10µs        ? ?/sec
filter context i32 high selectivity (kept 1023/1024)                          1.00      6.4±0.51µs        ? ?/sec    1.00      6.4±0.24µs        ? ?/sec
filter context i32 low selectivity (kept 1/1024)                              1.02    223.4±1.22ns        ? ?/sec    1.00    219.2±1.39ns        ? ?/sec
filter context i32 w NULLs (kept 1/2)                                         1.13     88.0±2.39µs        ? ?/sec    1.00     77.7±0.89µs        ? ?/sec
filter context i32 w NULLs high selectivity (kept 1023/1024)                  1.01      9.7±0.35µs        ? ?/sec    1.00      9.7±0.29µs        ? ?/sec
filter context i32 w NULLs low selectivity (kept 1/1024)                      1.00    443.4±2.20ns        ? ?/sec    1.00   443.1±14.13ns        ? ?/sec
filter context mixed string view (kept 1/2)                                   1.17    120.7±6.89µs        ? ?/sec    1.00    103.5±2.86µs        ? ?/sec
filter context mixed string view high selectivity (kept 1023/1024)            1.01     55.4±1.32µs        ? ?/sec    1.00     54.9±1.93µs        ? ?/sec
filter context mixed string view low selectivity (kept 1/1024)                1.00    622.4±4.78ns        ? ?/sec    1.00    625.2±5.69ns        ? ?/sec
filter context short string view (kept 1/2)                                   1.20    122.5±6.18µs        ? ?/sec    1.00    102.2±1.01µs        ? ?/sec
filter context short string view high selectivity (kept 1023/1024)            1.00     55.2±2.16µs        ? ?/sec    1.01     56.1±1.53µs        ? ?/sec
filter context short string view low selectivity (kept 1/1024)                1.00    497.4±4.61ns        ? ?/sec    1.00    496.9±4.46ns        ? ?/sec
filter context string (kept 1/2)                                              1.02   582.8±28.57µs        ? ?/sec    1.00    571.0±9.04µs        ? ?/sec
filter context string dictionary (kept 1/2)                                   1.00     16.9±0.06µs        ? ?/sec    1.00     16.9±0.35µs        ? ?/sec
filter context string dictionary high selectivity (kept 1023/1024)            1.05      7.3±0.29µs        ? ?/sec    1.00      6.9±0.30µs        ? ?/sec
filter context string dictionary low selectivity (kept 1/1024)                1.01    653.0±8.19ns        ? ?/sec    1.00    649.2±6.28ns        ? ?/sec
filter context string dictionary w NULLs (kept 1/2)                           1.13     88.5±0.60µs        ? ?/sec    1.00     78.7±2.79µs        ? ?/sec
filter context string dictionary w NULLs high selectivity (kept 1023/1024)    1.03     10.8±0.32µs        ? ?/sec    1.00     10.5±0.27µs        ? ?/sec
filter context string dictionary w NULLs low selectivity (kept 1/1024)        1.01   880.6±10.27ns        ? ?/sec    1.00    873.1±7.08ns        ? ?/sec
filter context string high selectivity (kept 1023/1024)                       1.00   669.0±13.30µs        ? ?/sec    1.02   685.1±16.07µs        ? ?/sec
filter context string low selectivity (kept 1/1024)                           1.00   881.3±16.42ns        ? ?/sec    1.01    888.4±3.91ns        ? ?/sec
filter context u8 (kept 1/2)                                                  1.00     14.9±0.11µs        ? ?/sec    1.00     15.0±0.13µs        ? ?/sec
filter context u8 high selectivity (kept 1023/1024)                           1.13      2.0±0.02µs        ? ?/sec    1.00  1798.1±12.55ns        ? ?/sec
filter context u8 low selectivity (kept 1/1024)                               1.01    217.8±0.75ns        ? ?/sec    1.00    216.3±0.91ns        ? ?/sec
filter context u8 w NULLs (kept 1/2)                                          1.14     86.2±0.92µs        ? ?/sec    1.00     75.9±0.35µs        ? ?/sec
filter context u8 w NULLs high selectivity (kept 1023/1024)                   1.00      5.1±0.03µs        ? ?/sec    1.02      5.2±0.02µs        ? ?/sec
filter context u8 w NULLs low selectivity (kept 1/1024)                       1.00    436.4±3.98ns        ? ?/sec    1.00    438.0±1.75ns        ? ?/sec
filter decimal128 (kept 1/2)                                                  1.00     42.2±5.21µs        ? ?/sec    1.36     57.2±2.79µs        ? ?/sec
filter decimal128 high selectivity (kept 1023/1024)                           1.00     51.3±1.75µs        ? ?/sec    1.03     52.9±1.54µs        ? ?/sec
filter decimal128 low selectivity (kept 1/1024)                               1.00  1836.0±15.36ns        ? ?/sec    1.61      3.0±0.03µs        ? ?/sec
filter f32 (kept 1/2)                                                         1.00    117.7±0.18µs        ? ?/sec    1.40    164.4±1.49µs        ? ?/sec
filter fsb with value length 20 (kept 1/2)                                    1.00    116.3±0.57µs        ? ?/sec    1.24    144.2±1.69µs        ? ?/sec
filter fsb with value length 20 high selectivity (kept 1023/1024)             1.00     69.0±2.78µs        ? ?/sec    1.03     71.3±2.97µs        ? ?/sec
filter fsb with value length 20 low selectivity (kept 1/1024)                 1.00      2.1±0.03µs        ? ?/sec    1.23      2.6±0.02µs        ? ?/sec
filter fsb with value length 5 (kept 1/2)                                     1.00    122.9±1.15µs        ? ?/sec    1.15    141.4±1.59µs        ? ?/sec
filter fsb with value length 5 high selectivity (kept 1023/1024)              1.00     10.2±0.60µs        ? ?/sec    1.08     11.1±0.57µs        ? ?/sec
filter fsb with value length 5 low selectivity (kept 1/1024)                  1.00      2.1±0.01µs        ? ?/sec    1.25      2.6±0.03µs        ? ?/sec
filter fsb with value length 50 (kept 1/2)                                    1.00   142.7±10.63µs        ? ?/sec    1.11    157.8±6.45µs        ? ?/sec
filter fsb with value length 50 high selectivity (kept 1023/1024)             1.03    211.8±5.70µs        ? ?/sec    1.00    204.7±5.36µs        ? ?/sec
filter fsb with value length 50 low selectivity (kept 1/1024)                 1.00      2.1±0.01µs        ? ?/sec    1.25      2.6±0.02µs        ? ?/sec
filter i32 (kept 1/2)                                                         1.00     29.4±0.09µs        ? ?/sec    1.82     53.6±0.06µs        ? ?/sec
filter i32 high selectivity (kept 1023/1024)                                  1.00      7.4±0.38µs        ? ?/sec    1.16      8.7±0.37µs        ? ?/sec
filter i32 low selectivity (kept 1/1024)                                      1.00  1783.6±11.65ns        ? ?/sec    1.64      2.9±0.02µs        ? ?/sec
filter optimize (kept 1/2)                                                    1.00     32.9±0.77µs        ? ?/sec    1.09     35.8±0.20µs        ? ?/sec
filter optimize high selectivity (kept 1023/1024)                             1.00      2.6±0.01µs        ? ?/sec    1.07      2.7±0.02µs        ? ?/sec
filter optimize low selectivity (kept 1/1024)                                 1.00   1496.9±3.60ns        ? ?/sec    1.69      2.5±0.01µs        ? ?/sec
filter run array (kept 1/2)                                                   1.02    456.1±1.69µs        ? ?/sec    1.00    449.3±7.12µs        ? ?/sec
filter run array high selectivity (kept 1023/1024)                            1.03    494.4±1.61µs        ? ?/sec    1.00    479.3±2.45µs        ? ?/sec
filter run array low selectivity (kept 1/1024)                                1.03    380.2±1.50µs        ? ?/sec    1.00   369.5±29.90µs        ? ?/sec
filter single record batch                                                    1.00     28.9±0.07µs        ? ?/sec    1.82     52.6±0.18µs        ? ?/sec
filter u8 (kept 1/2)                                                          1.00     28.8±0.05µs        ? ?/sec    1.25     36.0±0.18µs        ? ?/sec
filter u8 high selectivity (kept 1023/1024)                                   1.00      2.9±0.01µs        ? ?/sec    1.28      3.8±0.06µs        ? ?/sec
filter u8 low selectivity (kept 1/1024)                                       1.00  1834.0±12.41ns        ? ?/sec    1.62      3.0±0.05µs        ? ?/sec

@Dandandan
Copy link
Contributor Author

run benchmark arrow_reader_clickbench

@alamb-ghbot
Copy link

🤖 ./gh_compare_arrow.sh gh_compare_arrow.sh Running
Linux aal-dev 6.14.0-1018-gcp #19~24.04.1-Ubuntu SMP Wed Sep 24 23:23:09 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux
Comparing implement_fold (ea57375) to d6168e5 diff
BENCH_NAME=arrow_reader_clickbench
BENCH_COMMAND=cargo bench --features=arrow,async,test_common,experimental,object_store --bench arrow_reader_clickbench
BENCH_FILTER=
BENCH_BRANCH_NAME=implement_fold
Results will be posted here when complete

@alamb-ghbot
Copy link

🤖: Benchmark completed

Details

group                                             implement_fold                         main
-----                                             --------------                         ----
arrow_reader_clickbench/async/Q1                  1.00      2.3±0.01ms        ? ?/sec    1.01      2.3±0.01ms        ? ?/sec
arrow_reader_clickbench/async/Q10                 1.00     11.1±0.25ms        ? ?/sec    1.01     11.2±0.23ms        ? ?/sec
arrow_reader_clickbench/async/Q11                 1.00     12.9±0.38ms        ? ?/sec    1.00     12.9±0.29ms        ? ?/sec
arrow_reader_clickbench/async/Q12                 1.00     22.6±0.23ms        ? ?/sec    1.01     22.7±0.18ms        ? ?/sec
arrow_reader_clickbench/async/Q13                 1.00     28.1±0.24ms        ? ?/sec    1.00     28.0±0.24ms        ? ?/sec
arrow_reader_clickbench/async/Q14                 1.00     25.4±0.30ms        ? ?/sec    1.00     25.4±0.28ms        ? ?/sec
arrow_reader_clickbench/async/Q19                 1.00      5.5±0.13ms        ? ?/sec    1.02      5.6±0.14ms        ? ?/sec
arrow_reader_clickbench/async/Q20                 1.27    156.5±1.52ms        ? ?/sec    1.00   123.7±10.74ms        ? ?/sec
arrow_reader_clickbench/async/Q21                 1.01    164.2±0.85ms        ? ?/sec    1.00    163.1±1.30ms        ? ?/sec
arrow_reader_clickbench/async/Q22                 1.00   245.0±21.30ms        ? ?/sec    1.27   311.9±23.11ms        ? ?/sec
arrow_reader_clickbench/async/Q23                 1.01    407.1±2.19ms        ? ?/sec    1.00    403.1±9.80ms        ? ?/sec
arrow_reader_clickbench/async/Q24                 1.00     31.4±0.29ms        ? ?/sec    1.00     31.3±0.24ms        ? ?/sec
arrow_reader_clickbench/async/Q27                 1.02     99.6±0.66ms        ? ?/sec    1.00     98.0±0.73ms        ? ?/sec
arrow_reader_clickbench/async/Q28                 1.02     97.1±1.04ms        ? ?/sec    1.00     95.5±0.68ms        ? ?/sec
arrow_reader_clickbench/async/Q30                 1.01     28.1±0.36ms        ? ?/sec    1.00     27.8±0.25ms        ? ?/sec
arrow_reader_clickbench/async/Q36                 1.01    117.7±0.56ms        ? ?/sec    1.00    116.1±1.33ms        ? ?/sec
arrow_reader_clickbench/async/Q37                 1.02     93.0±0.43ms        ? ?/sec    1.00     91.4±0.70ms        ? ?/sec
arrow_reader_clickbench/async/Q38                 1.00     35.3±0.42ms        ? ?/sec    1.00     35.2±0.27ms        ? ?/sec
arrow_reader_clickbench/async/Q39                 1.02     46.0±0.44ms        ? ?/sec    1.00     45.3±0.28ms        ? ?/sec
arrow_reader_clickbench/async/Q40                 1.00     40.0±0.66ms        ? ?/sec    1.01     40.3±0.73ms        ? ?/sec
arrow_reader_clickbench/async/Q41                 1.00     29.3±0.18ms        ? ?/sec    1.01     29.5±0.37ms        ? ?/sec
arrow_reader_clickbench/async/Q42                 1.01     11.2±0.42ms        ? ?/sec    1.00     11.1±0.15ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q1     1.01      2.3±0.04ms        ? ?/sec    1.00      2.3±0.05ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q10    1.00     10.7±0.22ms        ? ?/sec    1.02     11.0±0.30ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q11    1.00     12.3±0.26ms        ? ?/sec    1.01     12.4±0.33ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q12    1.00     22.3±0.22ms        ? ?/sec    1.00     22.2±0.40ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q13    1.01     27.3±0.26ms        ? ?/sec    1.00     27.1±0.42ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q14    1.02     25.0±0.24ms        ? ?/sec    1.00     24.6±0.26ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q19    1.00      5.1±0.10ms        ? ?/sec    1.01      5.1±0.07ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q20    1.02    109.6±1.11ms        ? ?/sec    1.00    107.8±0.87ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q21    1.02    125.8±0.55ms        ? ?/sec    1.00    123.2±1.38ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q22    1.03    231.9±1.67ms        ? ?/sec    1.00    226.1±1.54ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q23    1.05    362.5±5.07ms        ? ?/sec    1.00    345.1±7.27ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q24    1.00     30.0±0.34ms        ? ?/sec    1.03     31.0±0.55ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q27    1.02     95.3±0.75ms        ? ?/sec    1.00     93.1±0.45ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q28    1.01     92.7±0.43ms        ? ?/sec    1.00     91.5±0.59ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q30    1.00     26.6±0.37ms        ? ?/sec    1.01     26.9±0.20ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q36    1.01    112.4±0.63ms        ? ?/sec    1.00    111.0±0.65ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q37    1.01     89.1±0.53ms        ? ?/sec    1.00     88.1±0.78ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q38    1.00     32.1±0.44ms        ? ?/sec    1.00     32.1±0.20ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q39    1.00     40.8±0.42ms        ? ?/sec    1.00     40.6±0.31ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q40    1.00     38.3±0.39ms        ? ?/sec    1.01     38.5±0.50ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q41    1.00     27.5±0.29ms        ? ?/sec    1.02     28.1±0.64ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q42    1.00     10.3±0.10ms        ? ?/sec    1.02     10.5±0.12ms        ? ?/sec
arrow_reader_clickbench/sync/Q1                   1.00  1970.0±30.31µs        ? ?/sec    1.01   1987.6±8.14µs        ? ?/sec
arrow_reader_clickbench/sync/Q10                  1.00      7.5±0.06ms        ? ?/sec    1.02      7.6±0.19ms        ? ?/sec
arrow_reader_clickbench/sync/Q11                  1.00      8.9±0.11ms        ? ?/sec    1.02      9.1±0.19ms        ? ?/sec
arrow_reader_clickbench/sync/Q12                  1.01     29.5±0.92ms        ? ?/sec    1.00     29.1±1.35ms        ? ?/sec
arrow_reader_clickbench/sync/Q13                  1.30     44.0±0.88ms        ? ?/sec    1.00     33.8±0.65ms        ? ?/sec
arrow_reader_clickbench/sync/Q14                  1.00     31.2±0.24ms        ? ?/sec    1.24     38.8±0.91ms        ? ?/sec
arrow_reader_clickbench/sync/Q19                  1.01      4.2±0.03ms        ? ?/sec    1.00      4.2±0.03ms        ? ?/sec
arrow_reader_clickbench/sync/Q20                  1.01    177.2±1.20ms        ? ?/sec    1.00    175.0±1.27ms        ? ?/sec
arrow_reader_clickbench/sync/Q21                  1.02    230.2±4.88ms        ? ?/sec    1.00    226.2±5.62ms        ? ?/sec
arrow_reader_clickbench/sync/Q22                  1.02    478.3±4.04ms        ? ?/sec    1.00    469.7±2.15ms        ? ?/sec
arrow_reader_clickbench/sync/Q23                  1.01   441.8±19.05ms        ? ?/sec    1.00   438.6±16.89ms        ? ?/sec
arrow_reader_clickbench/sync/Q24                  1.00     39.5±0.38ms        ? ?/sec    1.00     39.5±0.45ms        ? ?/sec
arrow_reader_clickbench/sync/Q27                  1.02    153.5±1.32ms        ? ?/sec    1.00    150.2±0.83ms        ? ?/sec
arrow_reader_clickbench/sync/Q28                  1.02    146.5±0.78ms        ? ?/sec    1.00    144.3±1.21ms        ? ?/sec
arrow_reader_clickbench/sync/Q30                  1.00     27.9±0.33ms        ? ?/sec    1.00     27.9±0.26ms        ? ?/sec
arrow_reader_clickbench/sync/Q36                  1.01    151.4±1.50ms        ? ?/sec    1.00    149.4±1.03ms        ? ?/sec
arrow_reader_clickbench/sync/Q37                  1.01     85.2±0.70ms        ? ?/sec    1.00     84.5±1.31ms        ? ?/sec
arrow_reader_clickbench/sync/Q38                  1.01     28.2±0.13ms        ? ?/sec    1.00     28.0±0.21ms        ? ?/sec
arrow_reader_clickbench/sync/Q39                  1.01     33.6±0.35ms        ? ?/sec    1.00     33.2±0.63ms        ? ?/sec
arrow_reader_clickbench/sync/Q40                  1.02     26.6±0.24ms        ? ?/sec    1.00     26.0±0.75ms        ? ?/sec
arrow_reader_clickbench/sync/Q41                  1.01     28.0±0.59ms        ? ?/sec    1.00     27.6±0.39ms        ? ?/sec
arrow_reader_clickbench/sync/Q42                  1.00     11.7±0.11ms        ? ?/sec    1.00     11.7±0.07ms        ? ?/sec

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

arrow Changes to the arrow crate

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants