Elasticsearch / Search
17 notes
Some notes are currently available only in Chinese. English translations are shown when available.
- Elasticsearch Transaction Loghistorical
The translog's role in write durability and shard recovery before Lucene changes are safely committed.
- How Elasticsearch Indexing Workshistorical
Lucene inverted indexes, terms, postings, doc values, segments, and the distinction between search and aggregation data structures.
- Elasticsearch CRUD Flowhistorical
How Elasticsearch routes document writes and reads through shards, replicas, refresh visibility, and durability mechanisms.
- Elasticsearch Architecturehistorical
Cluster nodes, indices, primary/replica shards, coordination, routing, distributed search, and recovery.
- Elasticsearch Performance Tuninghistorical
A workload-first approach to shard sizing, mappings, refresh, indexing, queries, caches, merges, and JVM/storage capacity.
- Elasticsearch Analyzershistorical
Character filters, tokenizers, token filters, index/search analyzers, and why analysis is part of the search schema.
- Elasticsearch Overviewhistorical
Distributed search and analytics built on Lucene: documents, indices, shards, replicas, inverted indexes, and near-real-time behavior.
- Elasticsearch Consistency Semanticshistorical
Primary-replica sequencing, acknowledgment, refresh visibility, optimistic concurrency, and limits of relational-style consistency assumptions.
- Synchronizing MySQL to Elasticsearchhistorical
Reliable change propagation from a relational source of truth into a search index using CDC/outbox/replayable pipelines.
- Using Elasticsearchhistorical
Core index, mapping, document, search, filter, aggregation, bulk, pagination, and alias practices.