SurgePilot

English

简体中文
日本語

English

简体中文
日本語

Appearance

SurgePilot DocsAI-Driven Distributed API Load Testing

Original baseline produced by AI agents. Design reusable request flows, run distributed load tests across your nodes, and inspect real-time metrics with zero engine friction.

Quickstart Guide
Startup Modes
Terms and FAQ
Illustrative SurgePilot AI load-testing artwork; not a performance benchmark

🤖 Original AI-Authored Baseline

The original baseline was produced by AI agents under human direction; the responsibility boundary and evidence map are documented for inspection.

⚡ One-Command Bootstrap

Run `surgepilot up` to interactively configure and launch the full stack — Web UI, FastAPI control plane, PostgreSQL, MinIO, Nginx, and InfluxDB/Grafana monitoring on host infrastructure in seconds.

🧾 Auditable Engineering Evidence

Governance documents, contracts, tests, and release assets keep the system inspectable from product intent through verification.

🌐 Distributed Linux Runners

Decoupled control plane and execution engines. Automate remote Linux Load Nodes over SSH/SFTP with multi-node metric aggregation and real-time InfluxDB metric streaming.

🔒 Data & Secret Isolation

Multi-team Workspace isolation, encrypted-at-rest Load Node credentials, masked Env Group secret variables, and contract-bound role permission checks.

📊 Authoritative Verdicts & Reports

Reusable request flows and load models deliver deterministic PASS/FAIL verdicts, detailed response metrics, failure breakdowns, and downloadable run artifacts.

What is SurgePilot? ​

SurgePilot is a self-hosted platform for designing, running, and reviewing API load tests. It separates reusable request flows from environment values and load settings, then executes them on one or more Linux Load Nodes.

SurgePilot is intended for developers, performance engineers, and internal platform teams that want a controlled load-testing workflow on infrastructure they operate.

Core workflow ​

SurgePilot separates what to test from how and where to run it:

  • Scenario — define reusable API and business request flows.
  • Env Group — supply environment-specific values without changing the flow.
  • Test Plan — combine a Scenario with load settings, execution resources, and pass/fail criteria.
  • Load Node — provide the Linux host that executes the test.
  • Run Report — review the verdict, metrics, diagnostics, and artifacts.

🔄 The Shortest Path ​

  1. Follow the Quickstart to start the complete stack with surgepilot up.
  2. Use First Run to create an environment, a Scenario, and a Test Plan.
  3. Attach a Load Node and execute your first controlled Run.
  4. Open Terms and FAQ when a product term or startup boundary is unclear.

Test only systems you are authorized to test

Even a small load test sends real traffic. Start with low concurrency and a short duration against a non-production target that you own or have explicit permission to test.