Search Systems & Elasticsearch

Inverted indexes, analyzers, BM25 relevance, autocomplete, fuzzy matching, CDC indexing pipelines, zero-downtime reindexing and cluster sizing.

About 120 minutes · 15 exam questions · 9 interview questions · 2 code labs

What you will learn

  1. Why your database is not a search engine
  2. The inverted index — build one yourself
  3. Analysis: tokenizers, filters, stemming and synonyms
  4. Mapping design: full-text vs keyword, multi-fields and filters
  5. Relevance: BM25 and how to tune ranking
  6. Autocomplete: edge n-grams, completion suggester, and a top-k trie
  7. Fuzzy search: Levenshtein distance and making it cheap
  8. The indexing pipeline: CDC → Kafka → indexer, and near-real-time refresh
  9. Zero-downtime reindexing with aliases
  10. Cluster architecture: sharding, replication and hot shards
  11. Deep pagination: why page 5,000 is expensive and what to do
  12. Scenario playbook: the seven classic search incidents

Next: Observability: Logs, Metrics, Traces & SLOs