update rsults

Author: MDA2AV

site/content/docs/add-framework/meta-json.md

@@ -13,7 +13,7 @@ Create a `meta.json` file in your framework directory:
   "description": "Short description of the framework and its key features.",
   "repo": "https://github.com/org/repo",
   "enabled": true,
-  "tests": ["baseline", "pipelined", "limited-conn", "json", "upload", "compression", "noisy", "baseline-h2", "static-h2"]
+  "tests": ["baseline", "pipelined", "limited-conn", "json", "upload", "compression", "noisy", "mixed", "baseline-h2", "static-h2"]
 }
 ```
 
@@ -41,6 +41,7 @@ Create a `meta.json` file in your framework directory:
 | `upload` | HTTP/1.1 | `/upload` |
 | `compression` | HTTP/1.1 | `/compression` |
 | `noisy` | HTTP/1.1 | `/baseline11` |
+| `mixed` | HTTP/1.1 | `/baseline11`, `/json`, `/db`, `/upload`, `/compression` |
 | `baseline-h2` | HTTP/2 | `/baseline2` (TLS, port 8443) |
 | `static-h2` | HTTP/2 | `/static/*` (TLS, port 8443) |
 | `baseline-h3` | HTTP/3 | `/baseline2` (QUIC, port 8443) |

site/content/docs/add-framework/test-profiles/h1/baseline.md

@@ -16,6 +16,38 @@ A mix of three request types, rotated across connections:
 
 This exercises the full HTTP handling path: request line parsing, header parsing, query string extraction, body reading (both Content-Length and chunked), integer arithmetic, and response serialization.
 
+## Expected request/response
+
+GET — sum of query parameters:
+
+```
+GET /baseline11?a=13&b=42 HTTP/1.1
+```
+
+```
+HTTP/1.1 200 OK
+Content-Type: text/plain
+
+55
+```
+
+POST with body — sum of query parameters + body:
+
+```
+POST /baseline11?a=13&b=42 HTTP/1.1
+Content-Type: text/plain
+Content-Length: 2
+
+20
+```
+
+```
+HTTP/1.1 200 OK
+Content-Type: text/plain
+
+75
+```
+
 ## What it measures
 
 - Raw request throughput under ideal conditions

site/content/docs/add-framework/test-profiles/h1/mixed.md

@@ -19,6 +19,29 @@ The Mixed Workload profile measures overall framework performance under a divers
 3. Each connection sends 100 requests of its assigned type, then disconnects and **reconnects with the next template type** (round-robin rotation)
 4. This rotation ensures all frameworks face a roughly even distribution of request types — fast connections cycle through all templates rather than staying on one type
 
+## Expected request/response
+
+Each template type uses the same endpoint format as its standalone profile:
+
+```
+GET /baseline11?a=1&b=2 HTTP/1.1
+→ 3
+
+GET /json HTTP/1.1
+→ {"items": [...], "count": 50}
+
+GET /db?min=10&max=50 HTTP/1.1
+→ {"items": [...], "count": N}
+
+POST /upload HTTP/1.1
+Content-Length: 1048576
+→ 1048576
+
+GET /compression HTTP/1.1
+Accept-Encoding: gzip
+→ (gzip-compressed JSON)
+```
+
 ## Template rotation
 
 Without rotation, connections assigned to lightweight endpoints (baseline: ~2 byte response) would complete hundreds of reconnect cycles, while heavy endpoints (compression: ~234 KB response) complete only a few. This would skew the results — frameworks efficient at heavy endpoints would paradoxically show lower total RPS.

site/content/docs/add-framework/test-profiles/h1/noisy.md

@@ -20,6 +20,31 @@ The Noisy profile measures how well a framework maintains throughput when valid
 
 4. Only **2xx responses** (from valid requests) count toward RPS — all error responses are ignored in scoring
 
+## Expected request/response
+
+Valid requests — same as baseline:
+
+```
+GET /baseline11?a=13&b=42 HTTP/1.1
+```
+
+```
+HTTP/1.1 200 OK
+Content-Type: text/plain
+
+55
+```
+
+Noise requests should return appropriate error codes:
+
+```
+GET /this/path/does/not/exist HTTP/1.1
+```
+
+```
+HTTP/1.1 404 Not Found
+```
+
 ## What it measures
 
 - **Error-handling overhead** — how much CPU the framework spends parsing and rejecting bad requests

site/content/docs/add-framework/test-profiles/h1/pipelined.md

@@ -6,6 +6,19 @@ title: Pipelined (16x)
 
 **Connections:** 512, 4,096, 16,384
 
+## Expected request/response
+
+```
+GET /pipeline HTTP/1.1
+```
+
+```
+HTTP/1.1 200 OK
+Content-Type: text/plain
+
+ok
+```
+
 ## What it measures
 
 - HTTP pipelining support and efficiency

site/content/docs/add-framework/test-profiles/h1/short-lived.md

@@ -6,6 +6,21 @@ Same workload as baseline, but each connection is closed and re-established afte
 
 **Connections:** 512, 4,096
 
+## Expected request/response
+
+Same as baseline — sum of query parameters:
+
+```
+GET /baseline11?a=13&b=42 HTTP/1.1
+```
+
+```
+HTTP/1.1 200 OK
+Content-Type: text/plain
+
+55
+```
+
 ## What it measures
 
 - Socket creation and teardown overhead

site/content/docs/add-framework/test-profiles/h2/baseline-h2.md

@@ -18,6 +18,19 @@ Same workload as the HTTP/1.1 baseline — query parameter parsing and sum compu
 - How frameworks handle many concurrent streams per connection
 - HPACK header compression performance
 
+## Expected request/response
+
+```
+GET /baseline2?a=1&b=1 HTTP/2
+```
+
+```
+HTTP/2 200 OK
+Content-Type: text/plain
+
+2
+```
+
 ## How it differs from baseline
 
 | | Baseline (HTTP/1.1) | Baseline (HTTP/2) |

site/content/docs/add-framework/test-profiles/h2/static-h2.md

@@ -29,6 +29,30 @@ Total payload: ~325 KB across 20 files.
 - Memory-mapped or pre-loaded file serving performance
 - TLS overhead with realistic mixed payloads
 
+## Expected request/response
+
+```
+GET /static/reset.css HTTP/2
+```
+
+```
+HTTP/2 200 OK
+Content-Type: text/css
+
+(file contents)
+```
+
+```
+GET /static/app.js HTTP/2
+```
+
+```
+HTTP/2 200 OK
+Content-Type: application/javascript
+
+(file contents)
+```
+
 ## How it differs from baseline-h2
 
 | | Baseline (HTTP/2) | Static Files (HTTP/2) |

site/content/docs/add-framework/test-profiles/h3/baseline-h3.md

@@ -19,6 +19,19 @@ The HTTP/3 Baseline profile tests raw throughput over QUIC, the UDP-based transp
 - **TLS 1.3 integration** — QUIC mandates encryption
 - **Framework QUIC implementation quality** — varies significantly across frameworks
 
+## Expected request/response
+
+```
+GET /baseline2?a=1&b=1 HTTP/3
+```
+
+```
+HTTP/3 200 OK
+Content-Type: text/plain
+
+2
+```
+
 ## Parameters
 
 | Parameter | Value |

site/content/docs/add-framework/test-profiles/h3/static-h3.md

@@ -33,6 +33,19 @@ The HTTP/3 Static Files profile serves 20 static files of various types over QUI
 | WebP | 3 | `hero.webp`, `thumb1.webp`, `thumb2.webp` |
 | JSON | 1 | `manifest.json` |
 
+## Expected request/response
+
+```
+GET /static/logo.svg HTTP/3
+```
+
+```
+HTTP/3 200 OK
+Content-Type: image/svg+xml
+
+(file contents)
+```
+
 ## Parameters
 
 | Parameter | Value |