Introducing Streamhouse: the open data architecture for AI | Learn More
Join us for a lightning talk about the experiences made by our team at RecruitCRM from offloading full-text search queries from MySQL to Elasticsearch, bringing query execution times from up to 100 seconds in the past down to less than five seconds with our new solution.
We’ll discuss why we decided to use change data capture (with Debezium), Kafka, and Flink for solving this task and dive into some of the challenges we made along the way: efficiently joining multiple change event streams, performant ingestion into Elasticsearch, creating representative benchmarks, and more. Benefit from our insights with building this real-time data pipeline, yielding not only a much better user experience, but also a more scalable and efficient system.