tokio
Engine · Rust
Minimal HTTP/1.1 server on raw tokio — no HTTP framework. A hand-rolled request parser (request line, headers, Content-Length + chunked bodies, keep-alive, pipelining, fragmented reads) on a tokio TcpStream, with one current-thread runtime per core and SO_REUSEPORT.
Composite rank
Each profile of a family is worth 100 to the entry that leads it, and the composite is the sum over the family. The field is this entry's own league: engines and reverse proxies are scored apart from frameworks. How it works.
| Family | Rank | Composite |
|---|---|---|
| HTTP/1.1 | 11 of 34 | 193 |
Every result
Alpha Round, 6 runs. Requests per second is the best of three; latency, CPU and memory come from that run.
| Category | Profile | Conns | Req/sec | Avg | p99 | CPU | Memory |
|---|---|---|---|---|---|---|---|
| Connection | Baseline | 512 | 4,269,664 | 119us | 266us | 6289.3% | 43MiB |
| Connection | Baseline | 4,096 | 4,187,085 | 977us | 1.34ms | 6417.9% | 187MiB |
| Connection | Pipelined | 512 | 54,824,854 | 148us | 272us | 6385.0% | 46MiB |
| Connection | Pipelined | 4,096 | 54,755,795 | 1.20ms | 1.62ms | 6057.7% | 164MiB |
| Connection | Short-lived | 512 | 2,563,668 | 185us | 638us | 5380.0% | 106MiB |
| Connection | Short-lived | 4,096 | 2,721,455 | 1.38ms | 2.36ms | 5727.1% | 426MiB |
How the benchmark is run · The machine · Add or fix an entry