Unleash Your Cluster's Potential with OpenSearch Benchmark
summary
OpenSearch Benchmark (v1.5) is a client tool that runs workloads like NYC taxis, HTTP logs, geopoint, vector search and the big five to measure latency, throughput and node telemetry; nightly runs persist metrics to an OpenSearch datastore and flag regressions (many queries under ~100 ms but some jumping to 300–400 ms). A custom-workflow example: a 5 TB, seven‑index production cluster on 15+ data nodes is modeled as a 1/5‑size workload where 1,500 queries are distilled to 15 and iterative OSB runs reproduce CPU and latency profiles.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 3:19What is OpenSearch Benchmark
- 12:34Tracking cluster performance
- 15:24Cluster performance issues
- 16:27Setting up OSB tests
- 17:52Customizing workloads
- 23:54Building custom workloads
- 29:52Future of OSB
related talks
23mWork Smarter, Not Harder: Discover Your OpenSearch Clusters Breaking...
41mScaling Search and Powering Vector-Based Retrieval With OpenSea...
19mEnhance Search Performance At Scale with
52mOperating OpenSearch at Scale: Lessons From Managing 10,000+ Clusters Across...
39mOperating OpenSearch at Scale With Kubernetes Operators: Lessons Learned and Community Insights
31mMaximize Resource Efficiency with Separated Index and Search Workloads
28mRunning OpenSearch at Scale in High-Traffic Gaming Systems
29mMastering Search Performance: Advanced Query Insights for Production OpenSearch