Lessons from Building Observability Tools at Netflix

  • Persistent log storage doesn’t scale. Work on an architecture that lets users stream logs into memory & keep the ones that match their criteria.
  • In a microservice architecture, knowing what decisions microservices are making & how those correlate is critical. Users need this additional context to find the root cause of issues that involve multiple systems.
  • Defining actionable alerting, beyond just threshold alerting, is key to system success at scale.

Full post here, 9 mins read