Versioned performance evidence
Rust edge performance you can audit before sizing servers.
Immutable benchmark snapshots tie every published result to a Webship release, test environment, and measurement date.
- Test platform
- Debian GNU/Linux 12 · Intel Xeon Gold 5115
- Accepted samples
- 5 × at least 20 seconds
- Reported unit
- Median requests per second
This 1.1.0 snapshot contains Webship-only measurements; use 1.0.0 for the retained contemporaneous competitor comparison. Reverse H1 TLS uses a 4/3/3 frontend/origin/client physical-core partition because the standard 4/4/2 run saturated its two-core TLS client. Other reverse TCP/TLS protocols use 4/4/2, and H3 uses 4/2/4.
Direct static response
Median requests per second. Higher is better.
| Server | HTTP/1.1 | HTTP/1.1 TLS | h2c | HTTP/2 TLS | HTTP/3 TLS |
|---|---|---|---|---|---|
| Webship | 415,735.58 | 312,069.33 | 1,041,848.3 | 937,318.6 | 307,727 |
Reverse proxy
Median requests per second. Higher is better.
| Server | HTTP/1.1 | HTTP/1.1 TLS | h2c | HTTP/2 TLS | HTTP/3 TLS |
|---|---|---|---|---|---|
| Webship | 164,919.51 | 120,834.25 | 127,965.65 | 112,687.2 | 98,897 |
Reproducible method
Each supported result is the median of five accepted samples. TCP and TLS workloads include warm-up; HTTP/3 includes a calibrated prime.
Server and load-generator processes ran on disjoint physical cores. Reverse-proxy runs also isolated the static origin.
Every accepted sample passed zero-error correctness gates and the published stability limits. Failed samples are never silently included.
A published snapshot is immutable. New Webship releases receive new benchmark versions instead of replacing historical results.