MariaDB Performance Audit
Database performance benchmarks with trend analysis and competitive context.
Competitive comparisons based on publicly available documentation as of June 2026. All SveltyCMS metrics self-measured via bun test tests/benchmarks/.
Console-only by default. Use BENCHMARK_RECORD=1 to write results to this report.
BENCHMARK_RECORD=1 bun test tests/benchmarks/auth-performance.test.tsThe full matrix runner (bun run scripts/benchmark-matrix/index.ts --sql) always records.
π Latest Performance Audit
π Fix Overlays
Phase notes and remediation entries appear here after matrix or surgical runs.
Partial update β dimension rollups and issues reflect only tests invoked in this run.
β PASS (partial invocation) Β· 1/41 tests Β· 40 skipped Β· mariadb Β· standalone run
Dimension health
| Dimension | Tests | Status | Worst Ξ | Issues |
|---|---|---|---|---|
| Core | 5 | π’ | +6099% | 0 |
| API | 0 | βͺ | β | 0 |
| Scale | 0 | βͺ | β | 0 |
| Resilience | 0 | βͺ | β | 0 |
Issues (action required)
β All clear β no regressions or budget violations detected.
Executive latency matrix
| Scenario | Latency | Trend | Budget | Result |
|---|---|---|---|---|
| Middleware | 0.074ms | βͺ stable at 0.085ms (Β±0.096ms) (3 runs) | < 5ms | π’ |
| Middleware | 0.808ms | π΄ slower: 0.085ms β 0.808ms (+851%) (3 runs) | < 5ms | π’ |
| Middleware | 1.048ms | π΄ slower: 0.085ms β 1.05ms (+1133%) (3 runs) | < 5ms | π’ |
| Middleware | 0.926ms | π΄ slower: 0.085ms β 0.926ms (+989%) (3 runs) | < 5ms | π’ |
| Middleware | 5.269ms | π΄ slower: 0.085ms β 5.27ms (+6099%) (3 runs) | < 5ms | π’ |
Historical pulse
Scope: Sparklines only β full run tables:
tests/benchmarks/results/history.sqlite
| Metric | Trend | Sparkline | Latest |
|---|---|---|---|
| cold start phased (cold) | π’ | β |
31976.981ms |
| database performance (warm) | π’ | βββ |
2.652ms |
π Latency trend chart (last 10 runs)
xychart-beta
title "mariadb REST p95 (last 10 runs)"
x-axis ["R1", "R2", "R3", "R4", "R5", "R6", "R7", "R8", "R9", "R10"]
y-axis "Latency (ms)"
line "p95" : [0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000, 0.000]
π Regression severity matrix (quadrant)
quadrantChart
title "Regression severity (impact vs delta)"
x-axis Low latency impact --> High latency impact
y-axis Low delta --> High delta
quadrant-1 Monitor
quadrant-2 Investigate soon
quadrant-3 Acceptable variance
quadrant-4 Critical β fix first
Middleware: [0.05, 0.05]
Middleware: [0.15, 0.14]
Middleware: [0.20, 0.19]
Middleware: [0.18, 0.16]
Middleware: [0.95, 0.95]
Current Run Summary (2026-07-07)
Scope: Only tests that ran in THIS invocation.
β οΈ Executive PASS/FAIL reflects the last full matrix run.
Run ID: f43331ff-473a-4ed3-9235-319efb91f320 Β· Tests invoked: 1 Β· Metrics: 5 Β· DB: mariadb
| Test | Metric | Avg (ms) | p95 (ms) | RPS | Trend | Detail |
|---|---|---|---|---|---|---|
| Hooks Performance | Static Asset (No Middleware) | 0.074 | 0.111 | 11842 | π’ | βͺ stable at 0.085ms (Β±0.096ms) (3 runs) |
| Hooks Performance | Turbo Pipeline (Light) | 0.808 | 1.268 | 1148 | βͺ | π΄ slower: 0.085ms β 0.808ms (+851%) (3 runs) |
| Hooks Performance | Full Security + Auth Pipeline | 1.048 | 1.472 | 954 | βͺ | π΄ slower: 0.085ms β 1.05ms (+1133%) (3 runs) |
| Hooks Performance | REST with API Caching | 0.926 | 1.344 | 1020 | βͺ | π΄ slower: 0.085ms β 0.926ms (+989%) (3 runs) |
| Hooks Performance | Mutation + Audit Logging | 5.269 | 6.187 | 179 | βͺ | π΄ slower: 0.085ms β 5.27ms (+6099%) (3 runs) |
Historical Trends (2026-07-07)
Scope: Sparklines only β full run tables live in
history.sqlite(link below). Runs = sparkline bar count.
Detail: tests/benchmarks/results/history.sqlite Β· SELECT test_id, phase, avg_ms, p95_ms, rps, timestamp FROM runs WHERE db_type='mariadb' AND redis=0 ORDER BY timestamp DESC
Series: 7 Β· DB: mariadb
| Metric | Trend | Sparkline (last 7) | Latest |
|---|---|---|---|
| CACHE PERFORMANCE (warm) | π’ | β |
0.385ms |
| COLD START PHASED (cold) | π’ | β |
31976.981ms |
| DATABASE PERFORMANCE (warm) | π’ | βββ |
2.652ms |
| HOOKS PERFORMANCE (warm) | π’ | ββββ |
0.074ms |
| REST API PERFORMANCE (warm) | π’ | β |
0.638ms |
| TRANSACTION ACID (warm) | π’ | β |
2.385ms |
| TRUTH LATENCY (warm) | π’ | β |
0.001ms |
π¬ Full benchmark ledger (41 modules)
Expand dimension groups, then any row, for ASCII truth tables. Executive summary above shows pass/fail and issues only.
Core Β· 18 tests
π·οΈ API LAYER LATENCY Β· π’ 5.592ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β API LAYER LATENCY β
β File: tests/benchmarks/api-latency.test.ts β
β Ran: 2026-07-26 10:33:35 β
β API Latency Benchmark (Production Optimized) β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β HTTP: findById @ 8c β 5.592 ms β p95: 9.823 ms β RPS: 175 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ PHASED COLD START Β· π’ 31976.981ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β PHASED COLD START AUDIT β
β File: tests/benchmarks/cold-start-phased.test.ts β
β Ran: 2026-07-07 13:27:56 β
β Measures server cold-start latency to READY state, using the built server. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Cold Start (IDLE β READY) β 31976.981 ms β p95: 38197.920 ms β RPS: 0 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ SRE TRUTH AUDIT Β· π’ 0.001ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β SRE TRUTH AUDIT β
β File: tests/benchmarks/truth-latency.test.ts β
β Ran: 2026-07-26 10:33:21 β
β Validates performance claims by comparing SDK, Middleware, and Real HTTP Stack β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Logic Baseline β 0.001 ms β p95: 0.002 ms β RPS: 1,626,898 β
β Local SDK (Full) β 0.008 ms β p95: 0.012 ms β RPS: 130,050 β
β HTTP End-to-End β 1.134 ms β p95: 2.402 ms β RPS: 882 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Incremental Content Reload Β· π’ 0.306ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β INCREMENTAL CONTENT RELOAD AUDIT β
β File: tests/benchmarks/content-incremental-reload.test.ts β
β Ran: 2026-07-21 16:13:30 β
β Measures surgical single-file fullReload vs full reconciliation path β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Incremental fullReload (1 file) β 0.306 ms β p95: 0.448 ms β RPS: 3,263 β
β Full Reconciliation Reload β 0.572 ms β p95: 0.804 ms β RPS: 1,747 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Middleware & Hooks Performance Β· π’ 1.212ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β MIDDLEWARE & HOOKS AUDIT β
β File: tests/benchmarks/hooks-performance.test.ts β
β Ran: 2026-07-26 10:36:36 β
βMeasures the cost of the full middleware chain including Turbo, Security, Auth, and Audit via HTTP E2E.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Static Asset (No Middleware) β 1.212 ms β p95: 2.144 ms β RPS: 807 β
β Turbo Pipeline (Light) β 0.880 ms β p95: 1.477 ms β RPS: 1,049 β
β Full Security + Auth Pipeline β 2.682 ms β p95: 4.019 ms β RPS: 371 β
β REST with API Caching β 3.874 ms β p95: 5.922 ms β RPS: 249 β
β Mutation + Audit Logging β 21.344 ms β p95: 27.703 ms β RPS: 46 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ State Machine Transitions Β· π’ 114.424ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β STATE MACHINE INTEGRITY β
β File: tests/benchmarks/state-machine-transition.test.ts β
β Ran: 2026-07-21 16:13:41 β
βSimulates rapid system re-initializations and verifies valid self-healing state transitions under stress.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β State Transition (READY -> IDLE -> READY) β 114.424 ms β p95: 138.293 ms β RPS: 9 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Database Adapter Raw CRUD Β· π’ 2.553ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β DATABASE ADAPTER PERFORMANCE (MARIADB) β
β File: tests/benchmarks/database-performance.test.ts β
β Ran: 2026-07-21 16:13:04 β
β Measures raw CRUD performance, indexing efficiency, and connection pool resilience β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β INSERT β 2.553 ms β p95: 4.373 ms β RPS: 350 β
β FIND ONE β 0.648 ms β p95: 0.903 ms β RPS: 1,304 β
β FIND MANY (limit 50) β 0.809 ms β p95: 1.130 ms β RPS: 1,067 β
β UPDATE β 1.342 ms β p95: 1.991 ms β RPS: 626 β
β NATIVE UPSERT β 0.581 ms β p95: 0.796 ms β RPS: 1,448 β
β COUNT β 0.773 ms β p95: 0.884 ms β RPS: 1,148 β
β DELETE β 1.595 ms β p95: 1.955 ms β RPS: 670 β
β BULK INSERT (100) β 8.159 ms β p95: 19.669 ms β RPS: 91 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ ACID Transaction Overhead Β· π’ 3.765ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β ACID INTEGRITY AUDIT β
β File: tests/benchmarks/transaction-acid.test.ts β
β Ran: 2026-07-21 16:13:09 β
β Measures transaction commit latencies and rollback overhead across database adapters β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β TX Commit β 3.765 ms β p95: 7.889 ms β RPS: 253 β
β TX Rollback Integrity β 3.255 ms β p95: 6.295 ms β RPS: 307 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Cache Β· π’ 2.375ms Β· β‘οΈ recorded (awaiting trend) Β· source
``` βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ β NEGATIVE CACHE PERFORMANCE AUDIT β β File: tests/benchmarks/negative-cache.test.ts β β Ran: 2026-07-26 10:35:21 β β Measures Bloom-filter style missing-key cache speedup for repeated misses β β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£ β First Miss (DB) β 2.375 ms β p95: 4.696 ms β RPS: 421 β β Cached Miss (Negative Cache) β 0.003 ms β p95: 0.005 ms β RPS: 360,081 β βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ ```π·οΈ Cache Hit/Miss Ratio & Efficiency Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Measures Redis cache hit/miss ratio, cold vs warm read speedup, and invalidation performance.
π·οΈ unknown Β· π’ 0.003ms Β· β‘οΈ recorded (awaiting trend) Β· source
``` βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ β SVELTYCMS β CACHE EVICTION BOUNDARY HARDENING β β File: unknown β β Ran: 2026-07-21 16:13:18 β β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£ β Cache Flooding & Eviction β 0.003 ms β p95: 0.005 ms β RPS: 299,244 β βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ ```π·οΈ Logic Β· π’ 0.001ms Β· β‘οΈ βͺ established at 0.001ms (1 run) (2026-07-07 13:28:45) Β· source
> β³ Pendingπ·οΈ SDK Β· π’ 0.933ms Β· β‘οΈ recorded (awaiting trend) Β· source
``` βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ β SVELTYCMS β SDK OVERHEAD TELEMETRY β β File: tests/benchmarks/local-api-performance.test.ts β β Ran: 2026-07-26 10:36:05 β β Measures LocalCMS SDK overhead vs direct adapter calls to verify zero-tax dispatching β β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£ β Direct Adapter Call β 0.933 ms β p95: 2.768 ms β RPS: 1,072 β β LocalCMS SDK Call β 0.724 ms β p95: 1.757 ms β RPS: 1,382 β βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ ```π·οΈ HTTP Β· π’ 0.580ms Β· β‘οΈ βͺ established at 0.580ms (1 run) (2026-07-07 13:28:45) Β· source
> β³ Pendingπ·οΈ Middleware Β· βͺ 5.269ms Β· βοΈ π΄ slower: 0.085ms β 5.27ms (+6099%) (3 runs) (2026-07-07 13:42:51) Β· source
> [!NOTE] > **Specific to this test**: Both avg and p95 degraded β likely adapter or infrastructure bottleneck. Check DB connection pool, indexes, or recent commits. Throughput dropped 98% β check connection pool saturation or lock contention. **Severe degradation** (+6099%) β review recent commits immediately.β³ Pending
π·οΈ System Β· π’ 0.638ms Β· β‘οΈ βͺ established at 0.638ms (1 run) (2026-07-07 13:32:29) Β· source
> β³ Pendingπ·οΈ Metadata Β· π’ 0.811ms Β· β‘οΈ βͺ established at 0.811ms (1 run) (2026-07-07 13:32:29) Β· source
> β³ Pendingπ·οΈ CRUD Β· π’ 1.193ms Β· β‘οΈ βͺ established at 1.19ms (1 run) (2026-07-07 13:32:29) Β· source
> β³ PendingAPI Β· 8 tests
π·οΈ Temporal Integrity Audit Β· π’ 13.485ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β TEMPORAL INTEGRITY β
β File: tests/benchmarks/temporal-integrity.test.ts β
β Ran: 2026-07-21 16:11:15 β
βValidates deterministic UTC normalization of ISO date strings across timezone offsets for consistent persistence.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Timezone Ingestion β 13.485 ms β p95: 13.485 ms β RPS: 74 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Relational & Nested Queries Β· π’ 6.243ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β RELATIONAL RESOLVER AUDIT β
β File: tests/benchmarks/relational-performance.test.ts β
β Ran: 2026-07-26 10:35:43 β
βMeasures GraphQL resolver performance for shallow and deep relational queries including joins and nested population.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Shallow Relational Query β 6.243 ms β p95: 9.486 ms β RPS: 152 β
β Deep Relational Query β 4.271 ms β p95: 6.416 ms β RPS: 221 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Auth & RBAC Performance Β· π’ 1.331ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β AUTHENTICATION TELEMETRY β
β File: tests/benchmarks/auth-performance.test.ts β
β Ran: 2026-07-26 10:33:40 β
β Authentication & RBAC Pipeline Benchmark (Production Optimized) β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Auth Validation @ 1c β 1.331 ms β p95: 1.966 ms β RPS: 602 β
β HTTP Auth Pipeline @ 8c β 3.377 ms β p95: 4.928 ms β RPS: 288 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ REST API Performance Β· π’ 0.896ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β ENTERPRISE REST API AUDIT β
β File: tests/benchmarks/rest-api-performance.test.ts β
β Ran: 2026-07-26 10:33:33 β
βMeasures latency, throughput, and correctness of core REST endpoints: health check, schema, CRUD list, and search.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Health Check β 0.896 ms β p95: 1.650 ms β RPS: 1,116 β
β Collection Schema β 2.281 ms β p95: 3.463 ms β RPS: 438 β
β Collection Find (List) β 2.903 ms β p95: 4.530 ms β RPS: 345 β
β Collection Search β 3.159 ms β p95: 4.654 ms β RPS: 317 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ GraphQL API Performance Β· π’ 1.549ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β GRAPHQL PERFORMANCE AUDIT β
β File: tests/benchmarks/graphql-api-performance.test.ts β
β Ran: 2026-07-26 10:34:58 β
β Measures GraphQL resolver performance and throughput across varied query scenarios. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β GQL: System Health β 1.549 ms β p95: 2.624 ms β RPS: 646 β
β GQL: Collection List β 6.226 ms β p95: 9.879 ms β RPS: 161 β
β GQL: Concurrent Load β 4.791 ms β p95: 8.347 ms β RPS: 209 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Enterprise SEO Suite Audit Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β E2E audit of Redirect Middleware, 404 Logging, and Sitemap Caching performance.
π·οΈ REAL-TIME BROADCAST AUDIT Β· π’ 0.202ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β YJS COLLABORATION SYNC AUDIT β
β File: tests/benchmarks/websocket-broadcast.test.ts β
β Ran: 2026-07-21 16:14:55 β
β Measures end-to-end Yjs update propagation latency and connection handshake timing. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Yjs CRDT Update Sync β 0.202 ms β p95: 0.291 ms β RPS: 4,946 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ WebSocket & Realtime Latency Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Benchmarks WebSocket connection/broadcast latency.
Scale Β· 18 tests
π·οΈ Negative Cache Performance Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Benchmarks 404-miss response times and cache lookup speedup.
π·οΈ Revision & History Growth Β· π’ 1.565ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β REVISION STRESS AUDIT β
β File: tests/benchmarks/revision-stress.test.ts β
β Ran: 2026-07-21 16:11:43 β
βMeasures read and history list latency degradation under heavy revision history growth (100 revisions per entry).β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Latest Read (High Revision Count) β 1.565 ms β p95: 2.467 ms β RPS: 639 β
β History List Retrieval β 0.900 ms β p95: 1.606 ms β RPS: 1,111 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Memory Stability & GC Behavior Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Long-running soak test to identify memory leaks and GC pressure.
π·οΈ Multi-Tenant Performance & Isolation Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Stress-tests cross-tenant isolation and security boundary latency.
π·οΈ Realistic Mixed Workload Β· π’ 4.167ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β MIXED WORKLOAD AUDIT β
β File: tests/benchmarks/mixed-workload.test.ts β
β Ran: 2026-07-26 10:35:52 β
βSimulates real-world traffic with a weighted mix of reads, searches, GraphQL queries, and metadata requests.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Mixed Workload β 4.167 ms β p95: 7.201 ms β RPS: 226 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ GraphQL Load Stress Β· π’ 6.288ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β GRAPHQL CAPACITY DISCOVERY β
β File: tests/benchmarks/graphql-stress.test.ts β
β Ran: 2026-07-26 10:35:08 β
β Discovers the server's maximum sustainable GraphQL throughput by ramping concurrency β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β GQL: 5c Warmup β 6.288 ms β p95: 8.834 ms β RPS: 144 β
β GQL: 10c Light β 9.995 ms β p95: 14.197 ms β RPS: 99 β
β GQL: 20c Moderate β 15.771 ms β p95: 20.790 ms β RPS: 63 β
β GQL: 40c Heavy β 30.067 ms β p95: 38.607 ms β RPS: 34 β
β GQL: 60c Stress β 43.164 ms β p95: 57.989 ms β RPS: 24 β
β GQL: 80c Extreme β 61.944 ms β p95: 84.819 ms β RPS: 16 β
β GQL: 100c Max β 49.548 ms β p95: 92.535 ms β RPS: 20 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Bulk Migration & Scale Β· π’ 57.451ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β MIGRATION & SCALE AUDIT β
β File: tests/benchmarks/migration-scale.test.ts β
β Ran: 2026-07-21 16:14:34 β
βMeasures bulk ingestion throughput for 10,000 entries and post-migration random lookup performance.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Bulk Migration (10k) β 57.451 ms β p95: 103.874 ms β RPS: 17 β
β Post-Migration Read β 4.036 ms β p95: 6.596 ms β RPS: 248 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Million-Row Index Pressure Β· π’ 4.523ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β INDEX PRESSURE AUDIT β
β File: tests/benchmarks/index-pressure.test.ts β
β Ran: 2026-07-21 16:11:36 β
β Measures read performance with sorting and filtering on a large entry collection. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Sorted List (25k rows) β 4.523 ms β p95: 7.385 ms β RPS: 221 β
β Filtered Query (25k rows) β 5.842 ms β p95: 9.155 ms β RPS: 171 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Content Scale Stress Β· π’ 1.329ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β CONTENT SCALE STRESS AUDIT β
β File: tests/benchmarks/content-scale-stress.test.ts β
β Ran: 2026-07-21 16:13:32 β
βMeasures file-scanning and content discovery performance at extreme scale (1,000+ collections)β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Cold Stress Scan (1k) β 1.329 ms β p95: 2.616 ms β RPS: 753 β
β Warm Stress Scan (1k) β 0.411 ms β p95: 0.556 ms β RPS: 2,432 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Media Upload Stress & Throughput Β· π’ 1.792ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β MEDIA UPLOAD STRESS AUDIT β
β File: tests/benchmarks/media-upload-stress.test.ts β
β Ran: 2026-07-21 16:14:19 β
βMeasures throughput for large file uploads, concurrent transfers, and streaming efficiency.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Single Upload (0.1MB) β 1.792 ms β p95: 2.653 ms β RPS: 558 β
β Small Upload (20KB) β 4.972 ms β p95: 8.229 ms β RPS: 201 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Full Client Journey Simulation Β· π’ 36.659ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β WORLD LIFE DATA β
β File: tests/benchmarks/client-journey.test.ts β
β Ran: 2026-07-26 10:41:41 β
βMeasures cumulative latency of a realistic editorial user workflow: Auth β List β View β Edit β Save.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Full Journey @ 4c β 36.659 ms β p95: 48.416 ms β RPS: 27 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Concurrency & Race Condition Β· βͺ β³ pending Β· β‘οΈ pending Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β THROUGHPUT SCALING (MARIADB) β
β File: tests/benchmarks/concurrency-throughput.test.ts β
β Ran: 2026-07-21 16:14:47 β
β 100 writes across 10, 100, 1000 docs. Same work, different parallelism. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β 10 docs Γ 10 β 0.000 ms β p95: 0.000 ms β RPS: 1,094 β
β 100 docs Γ 1 β 0.000 ms β p95: 0.000 ms β RPS: 770 β
β 1000 docs Γ 1 β 0.000 ms β p95: 0.000 ms β RPS: 966 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Multi-Doc Concurrency Throughput Β· π’ 0.360ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β LOCAL SDK BENCHMARK (MARIADB) β
β File: tests/benchmarks/local-api-throughput.test.ts β
β Ran: 2026-07-21 16:11:46 β
β Measures write throughput, read throughput, and SDK overhead in one pass. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Writes (1000) β 0.360 ms β p95: 0.000 ms β RPS: 2,781 β
β Reads (1000) β 0.198 ms β p95: 0.000 ms β RPS: 5,052 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Max Throughput β No Throttle Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β 1000 concurrent writes across 100 docs β no semaphore.
π·οΈ Local SDK Throughput Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β 1000 direct adapter calls β no HTTP, no middleware.
π·οΈ Failure Propagation & Fast-Fail Audit Β· π’ 0.001ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β AI INFRASTRUCTURE AUDIT β
β File: tests/benchmarks/ai-performance.test.ts β
β Ran: 2026-07-26 10:36:23 β
β Measures the latency tax of internal CMS logic for AI enrichment and layout generation. β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β AI Text Enrichment β 0.001 ms β p95: 0.002 ms β RPS: 793,966 β
β AI Layout Spec Generation β 0.000 ms β p95: 0.001 ms β RPS: 1,258,917 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Chaos & Infrastructure Resilience Β· π’ 197.326ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β CHAOS RESILIENCE AUDIT β
β File: tests/benchmarks/chaos-resilience.test.ts β
β Ran: 2026-07-26 10:34:44 β
βSimulates infrastructure brownouts and measures system stability and graceful degradation under stress.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Chaos Resilience (Brownout) β 197.326 ms β p95: 734.025 ms β RPS: 5 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ Production Day Simulation Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Full production day with mixed workload, 100% reliability.
Resilience Β· 5 tests
π·οΈ Circuit Breaker Failover Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Measures system graceful degradation when external services fail.
π·οΈ Throttling & Backoff Stress Β· π’ 2.555ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β THROTTLING AUDIT β
β File: tests/benchmarks/throttling-backoff-stress.test.ts β
β Ran: 2026-07-21 16:13:32 β
βSimulates high-velocity traffic from a single client IP to verify rate-limiting enforcement and backoff consistency.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β Throttling Enforcement β 2.555 ms β p95: 5.292 ms β RPS: 391 β
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββπ·οΈ GDPR Compliance Speed Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β Measures performance of deep-deletion across all linked tables.
π·οΈ Data Residency & Sovereignty Β· β³ pending Β· β‘οΈ baseline Β· source
β³ Pending β PII blocking and data sovereignty enforcement with cross-region failover.
π·οΈ Database Failover & Reconnection Β· π’ 18.143ms Β· β‘οΈ recorded (awaiting trend) Β· source
βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
β SVELTYCMS β DATA RESIDENCY AUDIT β
β File: tests/benchmarks/data-residency-failover.test.ts β
β Ran: 2026-07-26 10:36:42 β
βSimulates geopolitical boundary crossing and verifies PII field blocking for residency law compliance.β
β ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ£
β PII Field Blocking β 18.143 ms β p95: 22.896 ms β RPS: 55 β
ββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ³ Host environment populates after matrix run.
Was this page helpful?