How to Build and Monitor a Fast and Reliable Search Application
summary
Mondia’s search backend powers streaming and DTO services for millions of users (customers include o2 and Vodafone in Germany); the team implemented streaming indexing with Apache Flink and Apache Solr to make new items quickly available in the search index. They instrumented the architecture with extensive metrics to detect bottlenecks fast, and the talk walks through the system design, the observability tools used, and key learnings and pitfalls from their development experience.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 4:10User request flow
- 8:02Solr indexing pipeline
- 13:39Optimizing Solr collections
- 16:30Monitoring distributed systems
- 17:54Content delivery tracking
- 23:55DSL implementation tracking
- 29:51Summary
related talks
31mFluent Bit & OpenSearch: The Power Couple of Open Source Observability
11mLightning Talk: Federated Search Over Distributed Observability Data
33mStanding on the Shoulders of Giants
41mMonitoring Solr at Scale
46mCentralized logging and sleeping well at night
34mHow to Do Microservice Observability with OpenSearch
14mKey Considerations For Success When Deploying A Solution For Best-of-breed Observability At Scale
21mKubernetes Under the Magnifying Glass: Observability With OpenSearch