{
  "uid": "cogitave.learn.reliability-and-sre.summary",
  "kind": "moduleUnit",
  "href": "/modules/reliability-and-sre/summary/",
  "title": "Summary",
  "summary": "",
  "type": null,
  "products": [],
  "roles": [],
  "levels": [],
  "subjects": [],
  "headings": [
    "Next steps"
  ],
  "source": "You can now reason about a service's reliability targets the way the reliability\nstandard does, not just recite the word \"SLO.\"\n\nIn this module, you:\n\n- Learned the thesis that reliability is a **budgeted feature**, not an absolute,\n  and why that makes SRE an engineering discipline rather than a renamed ops\n  rota.\n- Set an SLO from a **critical user journey**, picked the right tier target, and\n  computed the **error budget** and **burn rate** it produces - 99.9% over 30\n  days buys about 43 minutes.\n- Read the **error-budget policy** - healthy vs. exhausted vs. repeated breach -\n  and named its one accountable owner, the reliability champion.\n- Saw what operating on that budget buys: sustainable on-call, a capped and\n  automated toil load, tested resilience through capacity and chaos, and an SLA\n  that is always looser than the SLO behind it.\n\n## Next steps\n\n- @cogitave.learn.respond-to-incidents - continue the **Operate the estate**\n  path here, where a burn-rate page becomes a declared incident under the\n  severity matrix.\n- The **reliability.md** standard, in the estate's standards repository, is\n  worth keeping open whenever you set or review a real service's SLO.\n- **ADR-0020** is the governance decision behind the error-budget policy.\n",
  "partOf": "cogitave.learn.reliability-and-sre",
  "durationInMinutes": 3,
  "quiz": null
}