github-actions[bot] commented on PR #7796: URL: https://github.com/apache/texera/pull/7796#issuecomment-5349149424
<!-- texera-benchmarks-comment --> ## ⚠️ Benchmark changes need a look 🟢 2 better · 🔴 4 worse · ⚪ 9 noise (<±5%) · 0 without baseline > Compared against main `ee06e4b` 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/32312116113) | | config | throughput | MB/s | latency | max Δ latest / 7d | |---|---|---:|---:|---:|---:| | 🔴 | bs=10 sw=10 sl=64 | 433 | 0.264 | 21,735/30,643/30,643 us | 🔴 +10.0% / 🔴 +96.5% | | 🟢 | bs=100 sw=10 sl=64 | 955 | 0.583 | 104,153/120,668/120,668 us | 🟢 -5.8% / 🔴 +13.0% | | ⚪ | bs=1000 sw=10 sl=64 | 1,109 | 0.677 | 898,265/999,996/999,996 us | ⚪ within ±5% / 🟢 -7.5% | <details><summary>Baseline details</summary> Latest main `ee06e4b` from same runner | config | metric | PR | latest main | 7d avg | Δ latest | Δ 7d | |---|---|---:|---:|---:|---:|---:| | bs=10 sw=10 sl=64 | throughput | 433 tuples/sec | 479 tuples/sec | 771.02 tuples/sec | -9.6% | -43.8% | | bs=10 sw=10 sl=64 | MB/s | 0.264 MB/s | 0.292 MB/s | 0.471 MB/s | -9.6% | -43.9% | | bs=10 sw=10 sl=64 | p50 | 21,735 us | 21,743 us | 12,749 us | -0.0% | +70.5% | | bs=10 sw=10 sl=64 | p95 | 30,643 us | 27,848 us | 15,594 us | +10.0% | +96.5% | | bs=10 sw=10 sl=64 | p99 | 30,643 us | 27,848 us | 19,320 us | +10.0% | +58.6% | | bs=100 sw=10 sl=64 | throughput | 955 tuples/sec | 976 tuples/sec | 997.05 tuples/sec | -2.2% | -4.2% | | bs=100 sw=10 sl=64 | MB/s | 0.583 MB/s | 0.596 MB/s | 0.609 MB/s | -2.2% | -4.2% | | bs=100 sw=10 sl=64 | p50 | 104,153 us | 101,566 us | 100,339 us | +2.5% | +3.8% | | bs=100 sw=10 sl=64 | p95 | 120,668 us | 128,099 us | 106,781 us | -5.8% | +13.0% | | bs=100 sw=10 sl=64 | p99 | 120,668 us | 128,099 us | 113,101 us | -5.8% | +6.7% | | bs=1000 sw=10 sl=64 | throughput | 1,109 tuples/sec | 1,093 tuples/sec | 1,037 tuples/sec | +1.5% | +6.9% | | bs=1000 sw=10 sl=64 | MB/s | 0.677 MB/s | 0.667 MB/s | 0.633 MB/s | +1.5% | +6.9% | | bs=1000 sw=10 sl=64 | p50 | 898,265 us | 918,148 us | 971,388 us | -2.2% | -7.5% | | bs=1000 sw=10 sl=64 | p95 | 999,996 us | 958,892 us | 1,013,249 us | +4.3% | -1.3% | | bs=1000 sw=10 sl=64 | p99 | 999,996 us | 958,892 us | 1,039,879 us | +4.3% | -3.8% | </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 0,10,10,64,20,461.79,200,128000,433,0.264,21735.29,30643.05,30643.05 1,100,10,64,20,2094.06,2000,1280000,955,0.583,104152.51,120667.70,120667.70 2,1000,10,64,20,18034.55,20000,12800000,1109,0.677,898265.13,999995.50,999995.50 ``` </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]
