Mentora
Classytic

Course

Cloud-Native Operations: package, schedule and operate a fleet

Designing a service and running one are different jobs. Replace drifting host installations with immutable images, hand a fleet a desired state and let controllers reconcile it, then judge the result by what customers actually experienced rather than by what the dashboards say.

56 lessons6h 6mEnglish

What you'll learn

  • Replace drifting host installation with a repeatable, digest-pinned container artifact
  • Distinguish container packaging from fleet scheduling and durable storage
  • Explain how service replicas, connection pooling and read replicas relieve different Supabase bottlenecks
  • Explain how controllers, schedulers and readiness checks reconcile desired state while finite capacity can leave Pods pending
  • Use bounded cells and federated clusters to contain blast radius
  • Define a customer-facing SLI and an SLO against it, rather than measuring what happens to be easy to collect
  • Use an error budget and a burn rate to decide whether something warrants waking a person
  • Choose rolling, canary or blue-green release mechanics, bound candidate exposure and roll back from customer-facing evidence
  • Sequence expand-and-contract data changes so old and new application versions can overlap safely
  • Give workloads attributable identities and task-scoped permissions that contain compromise blast radius
  • Rotate a shared secret by coordinating service acceptance, manager publication, consumer refresh and verified revocation
  • Distinguish configuration, state and actual infrastructure, then review and apply a plan that reconciles them deliberately

Requirements

  • Cloud Infrastructure Foundations, or equivalent understanding of VPC segmentation, load balancing, durable state and failure domains
  • Comfort with the idea that a service runs as a process on a machine somewhere. The packaging starts there

Course content

56 lessons · 6h 6m

About the creator

Classytic