github-actions[bot] commented on PR #8640: URL: https://github.com/apache/texera/pull/8640#issuecomment-6050661906
<!-- texera-benchmarks-comment --> ## ⚠️ Benchmark changes need a look 🟢 2 better · 🔴 2 worse · ⚪ 11 noise (<±5%) · 0 without baseline > Compared against main `40b507d` benchmarked on **this same runner**, so the delta is largely free of cross-runner hardware noise. The "7d avg" column still reflects the gh-pages dashboard. Treat <±5% as noise unless repeated. [Dashboard](https://apache.github.io/texera/dev/bench/) · [Run](https://github.com/apache/texera/actions/runs/37715510948) | | config | throughput | MB/s | latency | max Δ latest / 7d | |---|---|---:|---:|---:|---:| | 🟢 | bs=10 sw=10 sl=64 | 513 | 0.313 | 19,156/24,112/24,112 us | 🟢 -10.0% / 🔴 +77.7% | | 🔴 | bs=100 sw=10 sl=64 | 1,030 | 0.628 | 95,752/137,401/137,401 us | 🔴 +28.8% / 🔴 +46.3% | | ⚪ | bs=1000 sw=10 sl=64 | 1,197 | 0.731 | 831,698/886,680/886,680 us | ⚪ within ±5% / 🔴 -5.9% | <details><summary>Baseline details</summary> Latest main `40b507d` from same runner | config | metric | PR | latest main | 7d avg | Δ latest | Δ 7d | |---|---|---:|---:|---:|---:|---:| | bs=10 sw=10 sl=64 | throughput | 513 tuples/sec | 522 tuples/sec | 960.16 tuples/sec | -1.7% | -46.6% | | bs=10 sw=10 sl=64 | MB/s | 0.313 MB/s | 0.318 MB/s | 0.586 MB/s | -1.6% | -46.6% | | bs=10 sw=10 sl=64 | p50 | 19,156 us | 19,761 us | 11,059 us | -3.1% | +73.2% | | bs=10 sw=10 sl=64 | p95 | 24,112 us | 26,802 us | 13,567 us | -10.0% | +77.7% | | bs=10 sw=10 sl=64 | p99 | 24,112 us | 26,802 us | 16,824 us | -10.0% | +43.3% | | bs=100 sw=10 sl=64 | throughput | 1,030 tuples/sec | 1,062 tuples/sec | 1,235 tuples/sec | -3.0% | -16.6% | | bs=100 sw=10 sl=64 | MB/s | 0.628 MB/s | 0.648 MB/s | 0.754 MB/s | -3.1% | -16.7% | | bs=100 sw=10 sl=64 | p50 | 95,752 us | 93,601 us | 87,809 us | +2.3% | +9.0% | | bs=100 sw=10 sl=64 | p95 | 137,401 us | 106,694 us | 93,907 us | +28.8% | +46.3% | | bs=100 sw=10 sl=64 | p99 | 137,401 us | 106,694 us | 106,076 us | +28.8% | +29.5% | | bs=1000 sw=10 sl=64 | throughput | 1,197 tuples/sec | 1,201 tuples/sec | 1,272 tuples/sec | -0.3% | -5.9% | | bs=1000 sw=10 sl=64 | MB/s | 0.731 MB/s | 0.733 MB/s | 0.776 MB/s | -0.3% | -5.8% | | bs=1000 sw=10 sl=64 | p50 | 831,698 us | 834,736 us | 863,168 us | -0.4% | -3.6% | | bs=1000 sw=10 sl=64 | p95 | 886,680 us | 873,940 us | 907,700 us | +1.5% | -2.3% | | bs=1000 sw=10 sl=64 | p99 | 886,680 us | 873,940 us | 933,684 us | +1.5% | -5.0% | </details> <details><summary>Raw CSV</summary> ```csv config_idx,batch_size,schema_width,string_len,num_batches,total_ms,total_tuples,total_bytes,tuples_per_sec,mb_per_sec,lat_p50_us,lat_p95_us,lat_p99_us 0,10,10,64,20,389.56,200,128000,513,0.313,19156.48,24111.52,24111.52 1,100,10,64,20,1942.32,2000,1280000,1030,0.628,95752.43,137401.07,137401.07 2,1000,10,64,20,16709.23,20000,12800000,1197,0.731,831698.11,886679.74,886679.74 ``` </details> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
