Repository navigation
Benchmark integer statistics with distinct values counted - #10347
Conversation
Merging this PR will improve performance by 28.03%
|
| Mode | Benchmark | BASE |
HEAD |
Efficiency | |
|---|---|---|---|---|---|
| ⚡ | Simulation | integer_stats_nullable |
2.6 ms | 1.8 ms | +44.76% |
| ⚡ | WallTime | bitpack_blocked_compress_avx2 |
7.6 µs | 6.7 µs | +13.23% |
| 🆕 | Simulation | integer_stats_all_non_null |
N/A | 1.2 ms | N/A |
| 🆕 | Simulation | integer_stats_all_nullable |
N/A | 2.5 ms | N/A |
Tip
Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.
Comparing ji/int-stats-all-bench (69a3b50) with develop (d9ad4cf)
Footnotes
-
518 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩
The integer statistics benchmarks only covered the bounds and the runs. Add the same non-null and nullable arrays with distinct values counted, which computes every statistic, as when a dictionary scheme is enabled. Signed-off-by: Joe Isaacs <joe.isaacs@live.co.uk> Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_018J6qahqrVFb4UCPkwUzY11
0f5a7bf to
69a3b50
Compare
Summary
#10346 is stacked on this PR. It is independent of #10342, the other bench → impl stack (#10341 → #10342).
The
integer_statsCodSpeed benchmarks from #10341 only cover the bounds and the run count. When a dictionary scheme is enabled, the compressor also counts distinct values, so every statistic is computed. This adds that path, so CodSpeed tracks the distinct-counting changes in #10346.