Understanding ElasticSearch

- Introduction
- Mental model - Documents, Indices and shards
- Under the hood
- Mapping design

Transcript

1. Understanding ElasticSearch Devi A S L @ DAP sessions 6th Apr, 2022
2. Agenda • Introduction • Mental model - Documents, Indices and shards • Under the hood • Mapping design • Q & A
3. Introduction Distributed search engine, schema free, multi-tenant capable and based on Lucene library.
4. Use cases • Search Ranking • Text analysis and full text search • Aggregations • Highlighting, auto-correction, different languages, synonyms and many more.
5. Mental model Index - mapping - document - field - type
6. Mental model (Contd..)
7. What happens when we …
8. Start cluster • Configuration in elasticsearch.yml • Node discovery • Node roles and networking GET /_cat/nodes GET /_cat/shards GET /_cat/indices
9. What happens on adding a doc
10. Adding a doc • Finds a primary shard to routes the doc ◦ builds inverted indexes and other data structures • Replicates to replicas API: POST /<target>/_create/<_id>
11. Inverted index
12. Add document (Contd…)
13. What happens on index creation
15. What happens on search
16. Search = Query + Fetch
17. Distributed Fetch
18. Mapping Interesting Types • keyword • nested API: GET /<index>/_mapping
19. Mapping (Contd…) Interesting Attributes • index (should this be searchable?) • norms (should this be used for scoring?) • analyzer (should ES “analyze” this text?) • doc_values (should this be used for sorting/aggregations?) API: GET /<index>/_mapping
20. Q & A
21. References - https://www.elastic.co/guide/en/elasticsearch/reference/current/index.html - ElasticSearch terminology https://www.elastic.co/guide/en/elastic-stack-glossary/current/terms.html - Shard Allocation https://www.elastic.co/blog/every-shard-deserves-a-home - Defintive book https://allen8807.gitbooks.io/elasticsearch-definitive-guide-en/content/index.ht ml