Crate krabka_bench_driver 

Load driver and report aggregator for the Krabka vs Strimzi benchmark harness on Kubernetes.

Scenarios describe a target Kafka stack, a workload shape, and optional disturbance windows. The driver applies the scenario to Kubernetes, runs the producer/consumer load, samples Prometheus, and writes a RunOutput JSON artifact. The report binary reads those artifacts back and renders the side-by-side Markdown summary used in benchmark reports.

The crate ships two binaries:

• krabka-bench-driver — runs one scenario against one Kafka stack, either Krabka or Strimzi/Kafka. It captures throughput, latency, and disturbance data, queries Prometheus for resource usage, and writes a single RunOutput JSON file.
• krabka-bench-report — walks a directory of RunOutput files, groups them by scenario name, and writes a side-by-side Markdown summary.

Dimensioned values

Sizes, durations, and rates are [krabka_units] quantities throughout, and not bare numbers. A scenario’s msg_size is a ByteSize, its linger and duration are a Time, and a paced run’s rate is a Frequency. The operator writes them with units (512B, 5ms, 20000/s), and the measured RunOutput encodes them as exact integers. See scenario for the encoding of each field, and the code style guide for the vocabulary.

Command-line workflow

krabka-bench-driver \
  --scenario bench/scenarios/small-msg-saturate.yaml \
  --stack krabka \
  --namespace kafka-bench \
  --out runs/krabka-steady.json

krabka-bench-report --input runs --out report.md

Programmatic report aggregation

use {krabka_bench_driver::report, std::path::Path};

let markdown = report::render_markdown(Path::new("runs"), true)?;
std::fs::write("report.md", markdown)?;

Modules

aggregate

Cross-run aggregation, the “accurate averages” layer.

failover

Kill-broker orchestration for the failover scenario.

graph

Plotly HTML graph rendering for the benchmark report.

hist

HDR histogram helpers.

ids

Newtypes over the bare integers that the driver and report layers thread around.

payload

Record-payload generator.

prom

Prometheus query client.

qualification

Failover evidence checks and recovery analysis for benchmark runs.

rate

Token-bucket pacer for FixedRate scenarios. saturate mode never calls this. It only awaits the producer’s natural backpressure.

report

Aggregate per-run JSON outputs into a single Markdown summary.

scenario

Scenario is the on-disk YAML schema the driver reads. RunOutput is the on-disk JSON schema the driver writes. The report aggregator reads RunOutput documents and writes Markdown.

soak

The sustained-load and failure contract, and the judge that grades a run against it.

workload

Producer and consumer workload runners.