Skip to content

Commit 9c09ee0

Browse files
CybotTMCybotTM
committed
docs: update performance report with benchmark matrix data
- Update Rendertest metrics with accurate main branch data (8.16s, 97MB) - Add Resource Usage Comparison table with main vs feature memory data - Add Parallel Processing Comparison section (sequential, auto, 16 workers) - Update Cold vs Warm and Incremental sections with new measurements - Feature branch is 32% faster and uses 66% less memory even without parallelism
1 parent 18a91de commit 9c09ee0

File tree

1 file changed

+105
-24
lines changed

1 file changed

+105
-24
lines changed

‎docs/index.html‎

Lines changed: 105 additions & 24 deletions
Original file line numberDiff line numberDiff line change
@@ -462,37 +462,56 @@ <h3>Cold Build Times (No Cache)</h3>
462462
<tr>
463463
<td><strong>Rendertest</strong></td>
464464
<td>98</td>
465-
<td>11.24s</td>
466-
<td>5.6s</td>
467-
<td class="improvement">-50.2% (5.6s faster)</td>
465+
<td>8.16s</td>
466+
<td>5.53s</td>
467+
<td class="improvement">-32.2% (2.6s faster)</td>
468468
</tr>
469469
</tbody>
470470
</table>
471471

472-
<h3 style="margin-top: 2rem;">Resource Usage (Feature Branch)</h3>
472+
<h3 style="margin-top: 2rem;">Resource Usage Comparison</h3>
473473
<table class="comparison-table">
474474
<thead>
475475
<tr>
476476
<th>Documentation</th>
477+
<th>Branch</th>
477478
<th>CPU Usage</th>
478479
<th>Peak Memory</th>
479480
</tr>
480481
</thead>
481482
<tbody>
482483
<tr>
483-
<td><strong>TYPO3 Core Changelog</strong> (3666 files)</td>
484+
<td rowspan="2"><strong>TYPO3 Core Changelog</strong> (3666 files)</td>
485+
<td>main</td>
486+
<td>~100% (single-threaded)</td>
487+
<td>N/A</td>
488+
</tr>
489+
<tr>
490+
<td style="color: var(--primary); font-weight: 600;">feature</td>
484491
<td>437%</td>
485492
<td>942 MB</td>
486493
</tr>
487494
<tr>
488-
<td><strong>TYPO3 Core API</strong> (957 files)</td>
495+
<td rowspan="2"><strong>TYPO3 Core API</strong> (957 files)</td>
496+
<td>main</td>
497+
<td>~100% (single-threaded)</td>
498+
<td>N/A</td>
499+
</tr>
500+
<tr>
501+
<td style="color: var(--primary); font-weight: 600;">feature</td>
489502
<td>387%</td>
490503
<td>567 MB</td>
491504
</tr>
492505
<tr>
493-
<td><strong>Rendertest</strong> (98 files)</td>
494-
<td>188%</td>
495-
<td>120 MB</td>
506+
<td rowspan="2"><strong>Rendertest</strong> (98 files)</td>
507+
<td>main</td>
508+
<td>~59%</td>
509+
<td>97 MB</td>
510+
</tr>
511+
<tr>
512+
<td style="color: var(--primary); font-weight: 600;">feature</td>
513+
<td>200%</td>
514+
<td>33 MB</td>
496515
</tr>
497516
</tbody>
498517
</table>
@@ -501,9 +520,8 @@ <h3 style="margin-top: 2rem;">Resource Usage (Feature Branch)</h3>
501520
437% CPU = ~4.4 cores utilized on average.
502521
</p>
503522
<p style="color: var(--text-muted); margin-top: 0.5rem; font-size: 0.9rem;">
504-
<strong>Note:</strong> Main branch resource metrics not shown &mdash; the benchmark measurements for main were unreliable
505-
(identical values across runs, suggesting container limits rather than actual usage).
506-
Main branch is single-threaded, so CPU usage is ~100%.
523+
<strong>Memory Efficiency:</strong> Feature branch uses <strong>3x less memory</strong> for small docs (33 MB vs 97 MB)
524+
despite parallel processing. PHP profiling via <code>memory_get_peak_usage(true)</code>.
507525
</p>
508526

509527
<div class="chart-container">
@@ -533,11 +551,11 @@ <h4 style="margin-bottom: 1rem; text-align: center;">Build Time Comparison (seco
533551
</div>
534552
<div class="bar-group">
535553
<div style="display: flex; gap: 0.5rem; align-items: flex-end;">
536-
<div class="bar main" style="height: 3px;">
537-
<span class="bar-value">11.2s</span>
554+
<div class="bar main" style="height: 2px;">
555+
<span class="bar-value">8.16s</span>
538556
</div>
539557
<div class="bar" style="height: 1px;">
540-
<span class="bar-value">5.6s</span>
558+
<span class="bar-value">5.53s</span>
541559
</div>
542560
</div>
543561
<div class="bar-label">Rendertest<br>(98 files)</div>
@@ -593,17 +611,80 @@ <h2>Cold vs Warm Build Performance (Feature Branch)</h2>
593611
</tr>
594612
<tr>
595613
<td><strong>Rendertest</strong> (98 files)</td>
596-
<td>5.6s</td>
597-
<td>4.0s</td>
598-
<td>188%</td>
599-
<td>120 MB</td>
600-
<td><span style="color: var(--success); font-weight: bold;">-29%</span></td>
614+
<td>5.53s</td>
615+
<td>4.27s</td>
616+
<td>200%</td>
617+
<td>33 MB</td>
618+
<td><span style="color: var(--success); font-weight: bold;">-23%</span></td>
601619
</tr>
602620
</tbody>
603621
</table>
604622
<p style="color: var(--text-muted); margin-top: 1rem; font-size: 0.9rem;">
605623
<strong>Note:</strong> Warm builds benefit from Twig template caching persisted between runs.
606-
The 20-29% improvement comes primarily from avoiding Twig template recompilation.
624+
The 20-23% improvement comes primarily from avoiding Twig template recompilation.
625+
</p>
626+
</div>
627+
628+
<!-- Parallel Processing Comparison -->
629+
<div class="card">
630+
<h2>Parallel Processing Comparison (Rendertest - 98 files)</h2>
631+
<p style="color: var(--text-muted); margin-bottom: 1rem;">
632+
Comparison of different parallelism configurations on the small documentation set.
633+
</p>
634+
635+
<table class="comparison-table">
636+
<thead>
637+
<tr>
638+
<th>Configuration</th>
639+
<th>Cold Build</th>
640+
<th>Warm Build</th>
641+
<th>Memory</th>
642+
<th>CPU Usage</th>
643+
</tr>
644+
</thead>
645+
<tbody>
646+
<tr style="background: #f8f9fa;">
647+
<td><strong>main branch</strong> (official container)</td>
648+
<td>8.16s</td>
649+
<td>8.98s</td>
650+
<td>97 MB</td>
651+
<td>~59%</td>
652+
</tr>
653+
<tr>
654+
<td><strong>feature</strong> (sequential, --parallel-workers=-1)</td>
655+
<td>5.53s <span style="color: var(--success);">(-32%)</span></td>
656+
<td>4.27s <span style="color: var(--success);">(-52%)</span></td>
657+
<td>33 MB <span style="color: var(--success);">(-66%)</span></td>
658+
<td>200%</td>
659+
</tr>
660+
<tr>
661+
<td><strong>feature</strong> (auto, --parallel-workers=0)</td>
662+
<td>5.53s <span style="color: var(--success);">(-32%)</span></td>
663+
<td>4.25s <span style="color: var(--success);">(-53%)</span></td>
664+
<td>33 MB <span style="color: var(--success);">(-66%)</span></td>
665+
<td>195%</td>
666+
</tr>
667+
<tr>
668+
<td><strong>feature</strong> (16 workers, --parallel-workers=16)</td>
669+
<td>5.94s <span style="color: var(--success);">(-27%)</span></td>
670+
<td>4.43s <span style="color: var(--success);">(-51%)</span></td>
671+
<td>33 MB <span style="color: var(--success);">(-66%)</span></td>
672+
<td>196%</td>
673+
</tr>
674+
</tbody>
675+
</table>
676+
677+
<h3 style="margin-top: 1.5rem;">Key Observations</h3>
678+
<ul style="margin: 1rem 0; padding-left: 1.5rem;">
679+
<li><strong>32% faster even without parallelism:</strong> The feature branch's optimizations (caching patches, PHP 8.5 improvements) provide significant speedup independent of parallel processing.</li>
680+
<li><strong>66% less memory:</strong> Feature branch uses only 33 MB vs main's 97 MB &mdash; more efficient memory management.</li>
681+
<li><strong>Warm caching works:</strong> Feature branch warm builds are 23% faster than cold; main branch shows no warm cache benefit (8.98s warm vs 8.16s cold).</li>
682+
<li><strong>Parallelism overhead on small docs:</strong> For small documentation sets, forcing 16 workers adds slight overhead. Auto-detection or sequential mode perform similarly.</li>
683+
</ul>
684+
685+
<p style="color: var(--text-muted); margin-top: 1rem; font-size: 0.9rem;">
686+
<strong>Parallelism shines on large docs:</strong> The parallel processing benefits are most visible on larger documentation sets
687+
(Core API: 74% faster, Core Changelog: 96% faster) where the fork overhead is amortized across many files.
607688
</p>
608689
</div>
609690

@@ -980,9 +1061,9 @@ <h2>Incremental Build Performance (Feature Branch)</h2>
9801061
</tr>
9811062
<tr>
9821063
<td><strong>Rendertest</strong> (98 files)</td>
983-
<td>5.6s</td>
984-
<td>4.0s <span style="color: var(--success);">(-29%)</span></td>
985-
<td>4.1s <span style="color: var(--success);">(-27%)</span></td>
1064+
<td>5.53s</td>
1065+
<td>4.27s <span style="color: var(--success);">(-23%)</span></td>
1066+
<td>4.25s <span style="color: var(--success);">(-23%)</span></td>
9861067
</tr>
9871068
</tbody>
9881069
</table>

0 commit comments

Comments
 (0)