SLICK: Driving SLO Culture
summary
Meta builds SLICK as a central SLO tracking platform that indexes SLIs/SLOs, stores selected SLI time series for two years of full granularity, and exposes a GraphQL service plus CLI and integrations. A sharded MySQL backend and periodic pipelines persist per-minute SLI data. Adoption hits 1,000 services onboard and 80% of critical on-calls, Slack/Workplace bots surface regressions, and teams like LogDevice and a feed ranking service quantify regressions and recover to meet their objectives.
AI-written summary · about
chapters (9)
- 0:00SLO culture at Meta
- 2:10Historical SLO tracking
- 4:59Slick centralizes SLO discovery
- 7:26Long-term SLO data retention
- 8:51Slick architecture and workflows
- 12:21Slick data ingestion and storage
- 13:24Slick adoption and impact
- 16:37SLOs in Meta’s reliability
- 17:24Slick future directions
related talks
43mOur Experience Tracking and Driving SLO Adoption at Goldman Sachs
31mSLX: An Extended SLO Framework to Expedite Incident Recovery
37mReal World SLOs and SLIs: A Deep Dive
40mScaling SLOs with Kubernetes and Cloud Native Observability
19mImproving Reliability @ Meta: By Analyzing Historical Events That Led to SLO Violations
32m9 Things You Should Do When Starting to Use SLOs
37mUsing SLOs for Continuous Performance Optimizations of Your K8s Workloads
23mGoing from 30 to 30 Million SLOs