Speedup total_buffer_bytes_used

[Dandandan]

Which issue does this PR close?

• Closes #NNN.

Rationale for this change

This is hot for various clickbench queries in the benchmark.

What changes are included in this PR?

Are these changes tested?

Are there any user-facing changes?

[Dandandan]

run benchmark arrow_reader_clickbench

[alamb-ghbot]

🤖 ./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 speedup_total_buffer_bytes_used (0e2504b) 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=speedup_total_buffer_bytes_used
Results will be posted here when complete

[alamb-ghbot]

🤖: Benchmark completed

Details

group                                             main                                   speedup_total_buffer_bytes_used
-----                                             ----                                   -------------------------------
arrow_reader_clickbench/async/Q1                  1.00      2.3±0.02ms        ? ?/sec    1.00      2.3±0.03ms        ? ?/sec
arrow_reader_clickbench/async/Q10                 1.00     11.2±0.38ms        ? ?/sec    1.03     11.5±0.73ms        ? ?/sec
arrow_reader_clickbench/async/Q11                 1.00     12.8±0.39ms        ? ?/sec    1.00     12.9±0.33ms        ? ?/sec
arrow_reader_clickbench/async/Q12                 1.00     22.5±0.22ms        ? ?/sec    1.01     22.6±0.34ms        ? ?/sec
arrow_reader_clickbench/async/Q13                 1.00     27.8±0.28ms        ? ?/sec    1.01     28.0±0.52ms        ? ?/sec
arrow_reader_clickbench/async/Q14                 1.00     24.9±0.21ms        ? ?/sec    1.01     25.1±0.36ms        ? ?/sec
arrow_reader_clickbench/async/Q19                 1.00      5.5±0.07ms        ? ?/sec    1.02      5.6±0.15ms        ? ?/sec
arrow_reader_clickbench/async/Q20                 1.00    124.8±9.99ms        ? ?/sec    1.04    129.3±1.09ms        ? ?/sec
arrow_reader_clickbench/async/Q21                 1.00    162.5±1.19ms        ? ?/sec    1.00    162.3±0.95ms        ? ?/sec
arrow_reader_clickbench/async/Q22                 1.00    231.5±1.15ms        ? ?/sec    1.24   286.7±40.27ms        ? ?/sec
arrow_reader_clickbench/async/Q23                 1.00    400.1±2.36ms        ? ?/sec    1.00    401.0±2.93ms        ? ?/sec
arrow_reader_clickbench/async/Q24                 1.01     31.1±0.52ms        ? ?/sec    1.00     30.8±0.49ms        ? ?/sec
arrow_reader_clickbench/async/Q27                 1.00     96.9±0.46ms        ? ?/sec    1.00     96.8±1.07ms        ? ?/sec
arrow_reader_clickbench/async/Q28                 1.00     95.1±0.92ms        ? ?/sec    1.00     94.7±0.46ms        ? ?/sec
arrow_reader_clickbench/async/Q30                 1.00     27.6±0.36ms        ? ?/sec    1.00     27.6±0.26ms        ? ?/sec
arrow_reader_clickbench/async/Q36                 1.00    114.7±1.12ms        ? ?/sec    1.01    115.5±0.85ms        ? ?/sec
arrow_reader_clickbench/async/Q37                 1.00     90.8±0.63ms        ? ?/sec    1.00     90.8±0.72ms        ? ?/sec
arrow_reader_clickbench/async/Q38                 1.00     34.6±0.45ms        ? ?/sec    1.01     34.9±0.23ms        ? ?/sec
arrow_reader_clickbench/async/Q39                 1.00     45.0±0.64ms        ? ?/sec    1.00     45.2±0.34ms        ? ?/sec
arrow_reader_clickbench/async/Q40                 1.00     39.7±0.40ms        ? ?/sec    1.02     40.3±0.40ms        ? ?/sec
arrow_reader_clickbench/async/Q41                 1.00     28.9±0.31ms        ? ?/sec    1.02     29.5±0.42ms        ? ?/sec
arrow_reader_clickbench/async/Q42                 1.00     11.0±0.10ms        ? ?/sec    1.01     11.1±0.24ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q1     1.02      2.3±0.04ms        ? ?/sec    1.00      2.3±0.03ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q10    1.05     11.2±0.29ms        ? ?/sec    1.00     10.7±0.27ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q11    1.02     12.4±0.52ms        ? ?/sec    1.00     12.2±0.28ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q12    1.00     21.8±0.21ms        ? ?/sec    1.01     22.0±0.36ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q13    1.00     26.8±0.38ms        ? ?/sec    1.00     26.8±0.52ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q14    1.00     24.3±0.42ms        ? ?/sec    1.00     24.4±0.47ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q19    1.00      5.0±0.06ms        ? ?/sec    1.02      5.1±0.12ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q20    1.00    107.4±0.79ms        ? ?/sec    1.00    107.2±0.51ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q21    1.00    122.3±0.59ms        ? ?/sec    1.00    122.4±0.74ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q22    1.00    224.2±1.70ms        ? ?/sec    1.00    224.6±1.12ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q23    1.05   354.3±12.12ms        ? ?/sec    1.00    336.3±1.85ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q24    1.00     29.3±0.30ms        ? ?/sec    1.02     30.0±1.54ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q27    1.00     92.3±0.89ms        ? ?/sec    1.00     92.5±0.92ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q28    1.00     90.5±0.81ms        ? ?/sec    1.01     91.2±0.95ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q30    1.00     26.5±0.34ms        ? ?/sec    1.00     26.5±0.28ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q36    1.00    109.7±0.65ms        ? ?/sec    1.00    110.0±0.57ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q37    1.00     87.3±0.54ms        ? ?/sec    1.00     87.5±0.49ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q38    1.00     31.6±0.30ms        ? ?/sec    1.01     31.8±0.24ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q39    1.00     40.3±0.41ms        ? ?/sec    1.00     40.3±0.37ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q40    1.00     38.0±0.39ms        ? ?/sec    1.01     38.5±0.84ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q41    1.00     27.7±0.37ms        ? ?/sec    1.01     27.9±0.42ms        ? ?/sec
arrow_reader_clickbench/async_object_store/Q42    1.00     10.5±0.34ms        ? ?/sec    1.00     10.6±0.17ms        ? ?/sec
arrow_reader_clickbench/sync/Q1                   1.00   1987.9±7.05µs        ? ?/sec    1.00  1983.0±18.41µs        ? ?/sec
arrow_reader_clickbench/sync/Q10                  1.00      7.5±0.06ms        ? ?/sec    1.02      7.7±0.10ms        ? ?/sec
arrow_reader_clickbench/sync/Q11                  1.00      9.0±0.11ms        ? ?/sec    1.00      9.0±0.10ms        ? ?/sec
arrow_reader_clickbench/sync/Q12                  1.00     29.3±1.40ms        ? ?/sec    1.22     35.6±1.51ms        ? ?/sec
arrow_reader_clickbench/sync/Q13                  1.00     33.6±0.33ms        ? ?/sec    1.30     43.5±1.19ms        ? ?/sec
arrow_reader_clickbench/sync/Q14                  1.00     38.7±0.88ms        ? ?/sec    1.00     38.8±0.37ms        ? ?/sec
arrow_reader_clickbench/sync/Q19                  1.00      4.2±0.03ms        ? ?/sec    1.01      4.2±0.04ms        ? ?/sec
arrow_reader_clickbench/sync/Q20                  1.00    174.5±0.97ms        ? ?/sec    1.00    175.1±1.47ms        ? ?/sec
arrow_reader_clickbench/sync/Q21                  1.00    225.0±5.07ms        ? ?/sec    1.02    230.2±1.34ms        ? ?/sec
arrow_reader_clickbench/sync/Q22                  1.00    470.2±2.86ms        ? ?/sec    1.00    470.7±2.84ms        ? ?/sec
arrow_reader_clickbench/sync/Q23                  1.00   440.7±17.04ms        ? ?/sec    1.00   440.3±15.46ms        ? ?/sec
arrow_reader_clickbench/sync/Q24                  1.00     39.6±0.68ms        ? ?/sec    1.02     40.3±0.48ms        ? ?/sec
arrow_reader_clickbench/sync/Q27                  1.00    149.9±0.74ms        ? ?/sec    1.00    150.4±1.05ms        ? ?/sec
arrow_reader_clickbench/sync/Q28                  1.00    144.7±1.18ms        ? ?/sec    1.01    145.5±0.92ms        ? ?/sec
arrow_reader_clickbench/sync/Q30                  1.00     27.9±0.34ms        ? ?/sec    1.02     28.5±0.53ms        ? ?/sec
arrow_reader_clickbench/sync/Q36                  1.00    148.8±1.11ms        ? ?/sec    1.00    149.5±1.28ms        ? ?/sec
arrow_reader_clickbench/sync/Q37                  1.00     83.9±0.79ms        ? ?/sec    1.00     83.8±0.68ms        ? ?/sec
arrow_reader_clickbench/sync/Q38                  1.00     27.9±0.20ms        ? ?/sec    1.01     28.1±0.16ms        ? ?/sec
arrow_reader_clickbench/sync/Q39                  1.00     33.0±0.32ms        ? ?/sec    1.01     33.2±0.35ms        ? ?/sec
arrow_reader_clickbench/sync/Q40                  1.00     26.0±0.63ms        ? ?/sec    1.00     26.1±0.35ms        ? ?/sec
arrow_reader_clickbench/sync/Q41                  1.00     27.3±0.28ms        ? ?/sec    1.02     27.8±0.54ms        ? ?/sec
arrow_reader_clickbench/sync/Q42                  1.00     11.7±0.14ms        ? ?/sec    1.02     11.9±0.14ms        ? ?/sec