Dashboards, Not Drama: How Fixed-Cost Clouds Made O11y Fun Again
summary
A concrete claim: switching to a fixed-cost cloud model with platform-integrated observability cuts incident debugging from ~2 hours to 25 minutes vs ~2 hours by removing gated high-resolution data and delays. Evidence includes a $5,000 overnight bill spike, teams sampling metrics at 10% and deleting dashboards to save money, and one engineer spending 3 hours to save $200/month. The demo uses Cumulus with automatic collection and 30 days high-resolution data and predictable retention.
AI-written summary · about
chapters (8)
- 0:00Introduction and Welcome
- 0:19The Monitoring Fear Cycle
- 2:41Practical Impacts of Costly Monitoring
- 3:37Typical Incident Response Scenario
- 5:39Why Monitoring is Expensive
- 7:18A Better Alternative: Fixed Cost Cloud Model
- 7:49Real-World Benefits of Seamless Monitoring
- 9:15Conclusion and Final Thoughts
related talks
40mCloud Costs Panel "Lots of Reward Points"
46mWhen the CFO Comes Knocking: Our Journey to a Cost Efficient Data Platform
46mCloud cost chaos? Get clarity with FinOps cloud cost management
47mCost optimization
23mFinOps and Cloud Cost Optimization
30mNavigating Cloud Expenses in Data & AI: Strategies for Scientists and Engineers
43mBuilding Monetary Confidence in the Cloud
22mObservability is Too Expensive!