55 lines
1.9 KiB
Markdown
55 lines
1.9 KiB
Markdown
# Figures Index
|
|
|
|
Generated by:
|
|
|
|
```bash
|
|
.venv/bin/python analysis/characterization/plot_current_results.py
|
|
```
|
|
|
|
| Figure | Intended Claim |
|
|
|---|---|
|
|
| [fig_full_trace_workload.png](figures/fig_full_trace_workload.png) | Full GLM-5.1 trace is long-input, short-output, and high input/output ratio. |
|
|
| [fig_session_skew.png](figures/fig_session_skew.png) | Session input-token mass is highly skewed; top sessions dominate work. |
|
|
| [fig_pdsep_vs_combined.png](figures/fig_pdsep_vs_combined.png) | Existing static PD-sep A/B regresses TTFT/E2E vs combined. |
|
|
| [fig_elastic_vs_baseline.png](figures/fig_elastic_vs_baseline.png) | Existing elastic transfer-based run does not improve TTFT/TPOT over high-contention baseline. |
|
|
| [fig_gpu_balance.png](figures/fig_gpu_balance.png) | Existing runs show GPU-util imbalance, but more data is needed for hot-spot causality. |
|
|
| [fig_claim_status.png](figures/fig_claim_status.png) | Current audit separates supported, partial, and unsupported claims. |
|
|
|
|
## Figure Previews
|
|
|
|
### Full Trace Workload
|
|
|
|
Full GLM-5.1 trace is long-input, short-output, and high input/output ratio.
|
|
|
|
|
|
|
|
### Session Skew
|
|
|
|
Session input-token mass is highly skewed; top sessions dominate work.
|
|
|
|
|
|
|
|
### PD-Sep vs Combined
|
|
|
|
Existing static PD-sep A/B regresses TTFT/E2E vs combined.
|
|
|
|
|
|
|
|
### Elastic vs Baseline
|
|
|
|
Existing elastic transfer-based run does not improve TTFT/TPOT over high-contention baseline.
|
|
|
|
|
|
|
|
### GPU Balance
|
|
|
|
Existing runs show GPU-util imbalance, but more data is needed for hot-spot causality.
|
|
|
|
|
|
|
|
### Claim Status
|
|
|
|
Current audit separates supported, partial, and unsupported claims.
|
|
|
|
|