FinOps for Observability: Optimizing OpenSearch Costs Without Sacrificing Visibility
summary
Shows a working OpenSearch FinOps recipe that cuts a $15,200/month bill to save $4,100/month by combining ISM policies and rollover, hot/warm/cold tiering with prices 0.135 / 0.024 / 0.010 $ per GB-month, and data prep pipelines that drop debug logs and sample info at 10%. Implements one replica for hot and zero for warm/cold, aims for 10–50 GB shards, uses AWS ultra‑warm where available, and keeps 90‑day retention without losing alerts.
AI-written summary · about
chapters (15)
- 0:00FinOps and OpenSearch journey
- 2:45Observability cost challenges
- 4:15FinOps for cloud costs
- 5:23Log retention myths
- 6:28Tiered storage strategy
- 9:01Index State Management demo
- 12:19Cost visibility dashboards
- 13:21Filtering debug logs
- 14:45Shard size and replicas
- 15:53Cloud storage comparison
- 16:36Cost attribution policies
- 17:1730% cost reduction case
- 18:23FinOps maturity model
- 20:29Key takeaways
- 21:11Q&A
related talks
35mFrom Chaos To Clarity: Centralizing Observability With OpenSearch at Scale
35mDriving down the cost of observability
36mFinOps & Observability: technical implementation
13mShould an SRE Care About FinOps? Using Observability to...
15mThree key considerations for deploying best-of-breed observability
27mObservability is Too Damn Expensive!
22mObservability is Too Expensive!
10mWhy your observability is costing you too much?