Monitoring, logs and traces, how availability percentages convert into minutes of downtime, error budgets, and infrastructure as code.