Output backends¶

rampa ships metric samples to output backends during and after a test run. Use --output to send results to multiple destinations simultaneously. Local file backends create artifacts, remote backends deliver sample batches to external stores, and CI backends present results inside the workflow.

Built-in backends¶

Backend	Storage / delivery	Covers	Dependencies
`console`	Terminal summary (default)	Final human-readable run summary	None
`json`	Local JSON file	Sample artifact, final summary, threshold results	None
`csv`	Local CSV file	One row per sample for spreadsheets or warehouse import	None
`influxdb`	InfluxDB HTTP line protocol	Time-series storage	aiohttp (included)
`prometheus`	Prometheus remote write	Prometheus/Grafana metric storage	aiohttp; optional `python-snappy`
`otel`	OTLP/HTTP JSON	OpenTelemetry collector pipeline	aiohttp (included)
`github`	GitHub Actions annotations and step summary	CI presentation, not durable metric storage	None
`webhook`	HTTP JSON batches	Custom ingestion services	aiohttp (included)

Storage shape¶

Use json or csv when the run needs a durable file artifact. Use InfluxDB, Prometheus, OTEL, or webhook outputs when another system owns retention, dashboards, or downstream ingestion. Use the GitHub Actions surface for workflow summaries and annotations, usually alongside a JSON artifact for comparison.

CLI usage¶

$ rampa run load_test.py --output csv=results.csv

Multiple outputs in one run:

$ rampa run load_test.py \
  --output csv=results.csv \
  --output 'influxdb=http://localhost:8086/api/v2/write?org=myorg&bucket=rampa'

The --out flag is shorthand for --output json=<path>.

JSON¶

JSON is the default artifact format for CI comparison and machine-readable run storage.

$ rampa run load_test.py --out results.json

The file stores buffered samples. When the runner has a final metric snapshot, it also writes summary metrics and threshold results.

CSV¶

One row per sample. Tag keys become columns.

$ rampa run load_test.py --output csv=metrics.csv

Output:

timestamp,metric,value,method,scenario,status
1716691200,http_reqs,1.0,GET,smoke,200
1716691201,http_req_duration,45.2,GET,smoke,200

InfluxDB¶

Pushes samples as line protocol over HTTP.

$ rampa run load_test.py \
  --output 'influxdb=http://localhost:8086/api/v2/write?org=myorg&bucket=rampa'

Each sample becomes one line:

http_req_duration,method=GET,scenario=smoke value=45.2 1716691200000000000

Tags map to InfluxDB tags, the sample value maps to the value field, and the monotonic nanosecond timestamp maps to the InfluxDB timestamp.

Prometheus¶

Pushes metrics to Prometheus via the remote write API. Uses hand-crafted protobuf v1 encoding with snappy compression (falls back to gzip if python-snappy is not installed).

$ rampa run load_test.py \
  --output prometheus=http://localhost:9090/api/v1/write

Each sample becomes a Prometheus TimeSeries with __name__ set to the metric name and sample tags as labels. Feeds Grafana dashboards directly.

Install python-snappy for optimal compression:

$ uv add python-snappy

OpenTelemetry¶

Exports metrics via OTLP/HTTP+JSON to any OpenTelemetry-compatible collector (Grafana Alloy, OTEL Collector, Jaeger, etc.). Zero additional dependencies.

$ rampa run load_test.py --output otel=http://localhost:4318

The /v1/metrics path is appended automatically. Uses the JSON wire format (proto3 standard JSON mapping) — no protobuf compiler needed.

GitHub Actions¶

GitHub Actions output is for workflow presentation: threshold annotations and $GITHUB_STEP_SUMMARY content. Keep JSON or CSV enabled when the run also needs a downloadable artifact or a baseline for later comparison.

Webhook¶

POST sample batches as JSON to any HTTP endpoint.

$ rampa run load_test.py --output webhook=https://example.com/hook

Payload shape:

{
  "samples": [
    {"metric": "http_reqs", "value": 1.0, "timestamp": 1716691200, "tags": {"method": "GET"}}
  ]
}

Programmatic usage¶

Use get_output() to resolve backend names in programmatic runners.

from rampa.outputs import get_output

csv_out = get_output("csv", "results.csv")
influx_out = get_output("influxdb", "http://localhost:8086/api/v2/write")

All outputs implement the Output protocol: start(), add_samples(), and stop().