Scaling SLOs with Kubernetes and Cloud Native Observability
Talkat CloudOpen @ Open Source Summit Europe 2022 · on Thursday, 2022-09-15· at 11:00· Wicklow Hall 2A (Level 2), Convention Centre Dublin, Spencer Dock, N Wall Quay, North Wall Dublin 1, D01 T1·
summary
Claims a working SLO-as-code pipeline that generates Prometheus rules from OpenSLO/Sloth files, feeds metrics into Prometheus+Thanos+Grafana and alternate backends (New Relic, Datadog), and ties error budgets into CI so pipelines block or require enhanced approval when budgets burn. Implements synthetic tests, a Python observability API for contextual SLIs per customer/geo, and Terraform modules for dashboards/alerts. Dashboards and customer-impact reports exist as a mock-up; core automation and release gating are implemented and usable. OpenSLO+Sloth
AI-written summary · about
chapters (10)
- 0:00Start
- 6:30Reliability is CX
- 8:48What does that mean for the SLO process
- 10:47Common implementation issues
- 12:37Common scaling issues
- 15:22Cloud-native observability
- 17:23Observability as code
- 19:21Operationalizing
- 21:39SLOS a code - Sloth
- 23:32Observability Architecture
related talks
30mFrom Monitoring to Observability: Left Shift your SLOs with Chaos
31mServerless Observability: a case study of SLOs
35mFrom Monitoring to Observability: Left Shift your SLOs with Chaos
22mLook Ma, No Metrics: A Quest Towards Extracting SLOs
23mHow We Built a Complex SLO App Tightly Integrated with Grafana
25mAn SLO-Driven Approach to Enhance Kubernetes Cluster Reliability
34mSLO-Based Observability For All Kubernetes Cluster Components
47mService Level Objectives as Code: Bootstrapping SLOs with Terraform