github-actions[bot] commented on PR #8542: URL: https://github.com/apache/texera/pull/8542#issuecomment-5666196202
<!-- texera-benchmarks-comment --> ## ⚠️ Benchmark changes need a look 🟢 0 better · 🔴 11 worse · ⚪ 4 noise (<±5%) · 0 without baseline > Compared against main `5042d96` 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/34859248428) | | config | throughput | MB/s | latency | max Δ latest / 7d | |---|---|---:|---:|---:|---:| | 🔴 | bs=10 sw=10 sl=64 | 472 | 0.288 | 20,107/36,753/36,753 us | 🔴 +41.1% / 🔴 +140.0% | | 🔴 | bs=100 sw=10 sl=64 | 1,149 | 0.701 | 83,379/119,272/119,272 us | 🔴 +20.0% / 🔴 +16.3% | | 🔴 | bs=1000 sw=10 sl=64 | 1,382 | 0.844 | 712,841/849,959/849,959 us | 🔴 +8.6% / 🟢 +25.9% | <details><summary>Baseline details</summary> Latest main `5042d96` from same runner | config | metric | PR | latest main | 7d avg | Δ latest | Δ 7d | |---|---|---:|---:|---:|---:|---:| | bs=10 sw=10 sl=64 | throughput | 472 tuples/sec | 594 tuples/sec | 816.13 tuples/sec | -20.5% | -42.2% | | bs=10 sw=10 sl=64 | MB/s | 0.288 MB/s | 0.363 MB/s | 0.498 MB/s | -20.7% | -42.2% | | bs=10 sw=10 sl=64 | p50 | 20,107 us | 16,202 us | 12,250 us | +24.1% | +64.1% | | bs=10 sw=10 sl=64 | p95 | 36,753 us | 26,056 us | 15,315 us | +41.1% | +140.0% | | bs=10 sw=10 sl=64 | p99 | 36,753 us | 26,056 us | 19,159 us | +41.1% | +91.8% | | bs=100 sw=10 sl=64 | throughput | 1,149 tuples/sec | 1,224 tuples/sec | 1,067 tuples/sec | -6.1% | +7.7% | | bs=100 sw=10 sl=64 | MB/s | 0.701 MB/s | 0.747 MB/s | 0.651 MB/s | -6.2% | +7.7% | | bs=100 sw=10 sl=64 | p50 | 83,379 us | 80,396 us | 95,768 us | +3.7% | -12.9% | | bs=100 sw=10 sl=64 | p95 | 119,272 us | 99,396 us | 102,529 us | +20.0% | +16.3% | | bs=100 sw=10 sl=64 | p99 | 119,272 us | 99,396 us | 114,717 us | +20.0% | +4.0% | | bs=1000 sw=10 sl=64 | throughput | 1,382 tuples/sec | 1,384 tuples/sec | 1,098 tuples/sec | -0.1% | +25.9% | | bs=1000 sw=10 sl=64 | MB/s | 0.844 MB/s | 0.845 MB/s | 0.67 MB/s | -0.1% | +25.9% | | bs=1000 sw=10 sl=64 | p50 | 712,841 us | 721,102 us | 938,676 us | -1.1% | -24.1% | | bs=1000 sw=10 sl=64 | p95 | 849,959 us | 782,751 us | 979,813 us | +8.6% | -13.3% | | bs=1000 sw=10 sl=64 | p99 | 849,959 us | 782,751 us | 1,012,292 us | +8.6% | -16.0% | </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 0,10,10,64,20,423.48,200,128000,472,0.288,20106.78,36752.81,36752.81 1,100,10,64,20,1741.09,2000,1280000,1149,0.701,83378.71,119271.61,119271.61 2,1000,10,64,20,14466.69,20000,12800000,1382,0.844,712841.31,849959.44,849959.44 ``` </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]
