Distributed execution¶

Architecture¶

Coordinator (your machine)
  ├── MetricEngine (aggregated)
  ├── Threshold evaluation (centralized)
  └── WebSocket server
       ├── Worker 0 (SSH / Lambda / ECS)
       ├── Worker 1
       └── Worker N

Workers connect to the coordinator, receive work segments, run the test locally, and stream samples back.

Execution segments¶

Use ExecutionSegment to partition work deterministically across workers without central assignment:

from rampa.distributed.segment import ExecutionSegment

seg = ExecutionSegment(index=0, total=3)
seg.vu_range(30)       # range(0, 10)
seg.scale_rate(1000.0)  # 333.3

Each worker independently computes its share from its index and the total worker count.

Test archives¶

Self-contained .rampa zip bundles contain everything a remote worker needs. Archive creation is currently a programmatic API for coordinators and launchers, not a CLI command.

Contents: script, data files, requirements.txt, manifest.json. Archives are input bundles, not result stores. In a distributed run, the coordinator aggregates worker samples and output backends decide where results are retained: local JSON/CSV artifacts, remote metric stores, CI summaries, or custom ingestion.

For programmatic use, create_archive() builds the bundle and extract_archive() returns an ArchiveManifest when unpacking it:

from rampa.distributed.archive import create_archive, extract_archive

create_archive("load_test.py", "test.rampa", requirements=["aiohttp>=3.9"])
manifest = extract_archive("test.rampa", "worker-dir")

Wire protocol¶

Coordinator and workers communicate via WebSocket using MessagePack (JSON fallback). Message types: register, assign, samples, stop, heartbeat_req/resp, threshold_breach.