v0.1.2

HTTP load generator with coordinated-omission-corrected latency.

Binaries for Linux x86_64, macOS arm64 and Windows x86_64, each with a .sha256.

hammer -c 50 -t 4 -d 30 --rate 20000 --latency http://localhost:8080/

--rate measures latency from each request's intended departure. --closed-loop keeps the same schedule and measures from actual send, so both numbers come from one binary against one server. The README shows a run where p99 differs 48x between them.

Fixes since v0.1.1: the Windows binary reports its full version instead of truncating at the first dot.

v0.1.1

HTTP load generator with coordinated-omission-corrected latency.

Binaries for Linux x86_64, macOS arm64 and Windows x86_64, each with a .sha256.

hammer -c 50 -t 4 -d 30 --rate 20000 --latency http://localhost:8080/

--rate measures latency from each request's intended departure. --closed-loop keeps the same schedule and measures from actual send, so both numbers come from one binary on one server. The README shows a run where p99 differs 48x between them.

Fixes since v0.1.0: --version now prints the full tag rather than truncating at the first dot, and macOS is fully green in CI.

v0.1.0

HTTP load generator with coordinated-omission-corrected latency.

Binaries for Linux x86_64, macOS arm64 and Windows x86_64, each with a .sha256. GitHub adds the source archives.

npm install -g hammer-load
hammer -c 50 -t 4 -d 30 --rate 20000 --latency http://localhost:8080/

--rate measures latency from each request's intended departure. --closed-loop keeps the same schedule and measures from actual send, so you can compare the two on one server. See the README for a run where p99 differs 48x between them.