ES|QL FTW!
summary
Elasticsearch adds a piped SQL-like language called ES|QL with a new engine and endpoint "_query endpoint", shipped in 8.14 and requiring a modern Java client; it returns JSON but offers table/CSV formatting in Kibana and a selectable language: esql data view. Java integrations include a result-set esql adapter and an objects adapter that maps to Java records; ES|QL supports keep/drop fields, extract/compute functions, enrich-style joins, a 10,000 default limit, SIMD/Valhalla-driven speedups, and currently lacks vector-query parity.
AI-written summary · about
chapters (16)
- 0:00Introduction and setup
- 1:08About the speaker and ES|QL
- 2:11Elasticsearch and query languages
- 4:27What is ES|QL?
- 5:50ES|QL language basics
- 10:09Using ES|QL in Java
- 12:37Result handling in Java
- 14:00ES|QL in Kibana
- 16:52Response formats
- 19:19ES|QL engine and performance
- 20:44Performance engineering
- 22:55Visualization in Kibana
- 24:40Limitations and future
- 25:01Community and support
- 27:07Getting started with ES|QL
- 28:12Q&A session
related talks
34mEvolution of Slack Search
39mRelevant search results (with elasticsearch)
19mYou Know, for Search. Querying 24 Billion Records in 900ms
44mSearching through large graphs using Elasticsearch
14mSearch on Steroids
30mSearch on Steroids — A Story of Semantics
36mWhat’s New in the OpenSearch Project and Ecosystem
42mResearch in the Age of AI