Observability is Too Expensive - Chris Cooney
summary
Observability regularly eats 10–30% of cloud spend; indexing is the main cost and 99% of indexed logs are never searched. CorLogics rebuilds on a streaming engine, routes data into three tracks (frequent, monitoring, compliance) and offers direct-archive queries from customer S3, yielding reported 40–50% cost reductions. Practical tactics: gather usage stats (Prometheus, OpenSearch slow logs), transform verbose logs into metrics, push cold data to S3 and query with Athena or CorLogics’ archive query, and index cautiously with a use case driven storage plan.
AI-written summary · about
chapters (15)
- 0:00Introduction and speaker setup
- 1:07Why observability costs are skyrocketing
- 3:52Drivers of observability expense
- 6:02Data inefficiency in observability
- 9:55Common costly observability patterns
- 11:40Log storage tiering mistakes
- 13:46Use-case driven data storage
- 15:29Optimizing debug log costs
- 17:35Indexing is the cost killer
- 19:19Key question for SAS vendors
- 21:05Coralogix cost-saving features
- 23:52Design principles for cost control
- 25:16Q&A on high cardinality and filtering
- 27:42Q&A on reducing unnecessary logs
- 28:47Q&A on indexing-less solutions
related talks
35mDriving down the cost of observability
27mObservability is Too Damn Expensive!
22mObservability is Too Expensive!
10mWhy your observability is costing you too much?
15mThree key considerations for deploying best-of-breed observability
40mObservability that scales: Analyze data in stream and query remotely
31mMy Logs Cost More Than My Services
36mObservability? It's a Data Analytics Problem, You Fool!