ContextQMD
Back to Sglang

Ring SP Benchmark: Wan2.2-TI2V-5B (u1r2 vs Baseline)

docs_new/docs/sglang-diffusion/ring_sp_performance.mdx

0.5.113.6 KB
Original Source

This page reports Ring-SP performance for Wan2.2-TI2V-5B-Diffusers using:

  • Parallel config: sp=2, ulysses=1, ring=2 (short: u1r2)
  • Baseline config: sp=1, ulysses=1, ring=1 (short: u1r1)

Benchmark Setup

  • Model: Wan2.2-TI2V-5B-Diffusers
  • GPU: 48G RTX40 series * 2

Online Serving

Ring SP (u1r2)

bash
sglang serve \
  --model-type diffusion \
  --model-path /model/HuggingFace/Wan-AI/Wan2.2-TI2V-5B-Diffusers \
  --num-gpus 2 --sp-degree 2 --ulysses-degree 1 --ring-degree 2 \
  --port 8898

Baseline (u1r1)

bash
sglang serve \
  --model-type diffusion \
  --model-path /model/HuggingFace/Wan-AI/Wan2.2-TI2V-5B-Diffusers \
  --num-gpus 1 --sp-degree 1 --ulysses-degree 1 --ring-degree 1 \
  --port 8898

Benchmarks

Benchmark Disclaimer

These benchmarks are provided for reference under one specific setup and command configuration. Actual performance may vary with model settings, runtime environment, and request patterns.

Stage Time Breakdown

<table style={{width: "100%", borderCollapse: "collapse", tableLayout: "fixed"}}> <colgroup> <col style={{width: "25%"}} /> <col style={{width: "25%"}} /> <col style={{width: "25%"}} /> <col style={{width: "25%"}} /> </colgroup> <thead> <tr> <th>Stage / Metric</th> <th><code>u1r2</code> (s)</th> <th><code>u1r1</code> baseline (s)</th> <th>Speedup</th> </tr> </thead> <tbody> <tr> <td>InputValidation</td> <td>0.1060</td> <td>0.1029</td> <td>0.97x</td> </tr> <tr> <td>TextEncoding</td> <td>1.3965</td> <td>2.2261</td> <td>1.59x</td> </tr> <tr> <td>LatentPreparation</td> <td>0.0002</td> <td>0.0002</td> <td>1.00x</td> </tr> <tr> <td>TimestepPreparation</td> <td>0.0003</td> <td>0.0004</td> <td>1.33x</td> </tr> <tr> <td>Denoising</td> <td>52.6358</td> <td>71.6785</td> <td>1.36x</td> </tr> <tr> <td>Decoding</td> <td>7.6708</td> <td>13.4314</td> <td>1.75x</td> </tr> <tr> <td><strong>Total</strong></td> <td><strong>63.74</strong></td> <td><strong>90.63</strong></td> <td><strong>1.42x</strong></td> </tr> </tbody> </table>

Memory Usage

<table style={{width: "100%", borderCollapse: "collapse", tableLayout: "fixed"}}> <colgroup> <col style={{width: "25%"}} /> <col style={{width: "25%"}} /> <col style={{width: "25%"}} /> <col style={{width: "25%"}} /> </colgroup> <thead> <tr> <th>Memory Metric</th> <th><code>u1r2</code> (GB)</th> <th><code>u1r1</code> baseline (GB)</th> <th>Delta</th> </tr> </thead> <tbody> <tr> <td>Peak GPU Memory</td> <td>20.07</td> <td>27.40</td> <td>-7.33</td> </tr> <tr> <td>Peak Allocated</td> <td>13.35</td> <td>20.40</td> <td>-7.05</td> </tr> <tr> <td>Memory Overhead</td> <td>6.72</td> <td>7.00</td> <td>-0.28</td> </tr> <tr> <td>Overhead Ratio</td> <td>33.5%</td> <td>25.6%</td> <td>+7.9pp</td> </tr> </tbody> </table>

Summary

  • End-to-end latency improves from 90.63s to 63.74s (1.42x).
  • Main gains come from Denoising (1.36x) and Decoding (1.75x).
  • Absolute memory usage drops noticeably on Ring-SP (Peak GPU Memory -7.33GB, Peak Allocated -7.05GB).
  • Overhead ratio rises (+7.9pp), so future tuning can focus on reducing communication/runtime overhead while preserving the latency gain.