Benchmarks

Real, measured performance — not theoretical.

Test Setup

• Machine: MacBook Pro (Linux/macOS)
• Tool: wrk -t8 -c400 -d10s
• Build: cargo build --release
• Endpoint: /health

Results

TezWeb Across Modes

Logger middleware adds noticeable overhead — for max throughput, disable it or use sampling in production.

Reproducing the Benchmark

cd tezweb
cargo build --release --example rest_api
./target/release/examples/rest_api &
wrk -t8 -c400 -d10s http://localhost:8080/health

Actix Comparison

cd actix-bench
cargo build --release
./target/release/actix-bench &
wrk -t8 -c400 -d10s http://localhost:8081/

Why TezWeb Is Fast

1. Hand-rolled HTTP/1.1 parser — no Hyper overhead for HTTP/1.1
2. Custom Trie router — O(log n) lookup
3. Per-core TCP listeners with SO_REUSEPORT
4. Thread-local buffer pooling — reduces allocations
5. Zero unnecessary middleware by default

Platform Notes

• Linux: Uses monoio + io_uring — expected to outperform macOS numbers shown here
• macOS: kqueue via tokio — current benchmark numbers (40,248 RPS)
• Windows: IOCP via tokio — untested