feedback              the second way

telemetry
  • create centralized telemetry infrastructure
  • create application logging telemetry
  • use telemetry to guide troubleshooting
  • enable creation of production metrics as part of daily work
  • create self-service access to to telemetry and information radiators
  • application metrics & business metrics
  • infrastructure metrics
  • overlay other information onto available metrics
analyze
  • use mean & standard variation to detect potential problems
  • instrument & alert on undesired outcomes
enable feedback so dev & ops can safely deploy code
  • use telemetry to make deployments safer
  • dev shares pager duty with ops
  • devs follow work downstream
  • devs self-manage their production service (initially
review & co-ordinate processes
  • github flow
  • change approval processes
  • enable peer reviews
  • enable pair programming
  • evaluate pull request process
  • cut bureaucratic processes - fearlessly