🗓️ 05102024 1201

ES REALTIME SEARCH

Terms

Term	Description
lucene index	underlying data structure that Elasticsearch uses to store and retrieve information
segment	smaller part of a Lucene index As documents are indexed, Elasticsearch splits data into multiple segments.
commit point	snapshot of all the segments at a given point in time It marks the point where data has been committed and is now searchable
elasticsearch refresh	refresh process in Elasticsearch ensures that newly indexed data becomes visible to search

How Elasticsearch Real-Time Search Works

1. In-memory buffer: As new documents are indexed, they are first added to an in-memory buffer.
2. Commit Point Creation: At certain intervals or upon manual triggering, Elasticsearch writes the in-memory buffer to disk, creating new segments. At this point, a commit point is created.
3. Refresh Process: After a refresh, the new segments are opened, and the indexed data becomes searchable.

---

References

• https://www.elastic.co/guide/en/elasticsearch/reference/current/near-real-time.html