Unified Performance Observability in OpenSearch: Integrating Workload Management and Query Insights
summary
Query Insights records latency CPU and memory by search phase and treats WLM group info as a first class attribute so queries are filterable and grouped by shape via OpenSearch Dashboards and local index exporters. Rule based auto tagging assigns requests by index username or role and WLM enforces limits proactively with rejections and reactively by canceling heavy queries. In the demo nested aggregations on an analytics index show 5+ second CPU spikes that drop after applying 15% CPU cap, using grouping by query shape and rule-based auto tagging.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:44Query Insights in OpenSearch
- 3:54Why Query Insights matters
- 5:03Workload Management intro
- 11:28High-resource queries analysis
- 16:21Workload groups setup
- 18:50Summary
related talks
17mMastering the Query Insights Dashboard in OpenSearch
38mQuery Insights Deep Dive: From Architecture to Production Optimization
15mFrom Top N Lists to Visual Analytics: Query Insights Dashboard in OpenSearch
23mBeyond Slow Logs: Implementing Advanced Query Insights in OpenSearch
04mOpenSearch Performance and Monitoring
29mThe osquery Agent. Performance Tuning @Scale
26mPlug-and-Play Observability: OTel Auto-Instrumentation for LLMs in OpenSearch
31mMaximize Resource Efficiency with Separated Index and Search Workloads