Mastering the Query Insights Dashboard in OpenSearch
summary
Query Insights captures per-request latency CPU and memory, buffers completed searches and drains into three priority queues every 5 seconds, keeping the top N (default 10) per metric for a rolling window (default 5 minutes), then exports to one of three sinks (default local index), with local indices retained 7 days by default (configurable to 180). Dashboard shows phase-by-phase latency, coordinator node, grouping of similar queries, live query listing and cancellation, and APIs for top queries live queries and health stats.
AI-written summary · about
chapters (9)
- 0:00Query Insights intro
- 2:53Features and config
- 5:42Query details explained
- 6:47Live queries insights
- 7:10Query Insights APIs
- 8:16Top queries tracking
- 10:02Exporting query data
- 11:31Future roadmap
- 13:40Resources & Q&A
related talks
15mFrom Top N Lists to Visual Analytics: Query Insights Dashboard in OpenSearch
38mQuery Insights Deep Dive: From Architecture to Production Optimization
23mBeyond Slow Logs: Implementing Advanced Query Insights in OpenSearch
20mUnified Performance Observability in OpenSearch: Integrating Workload Management and Query Insights
20mMeasuring Data with OpenSearch to Better Serve the Community
18mEnhance Query Execution Visibility in OS with OpenTelemetry
04mOpenSearch Performance and Monitoring
29mThe osquery Agent. Performance Tuning @Scale