ReadMeFeed / Community knowledge
Real questions. Useful conversations. Find the people who know your stack.
A decision guide for selecting Neo4j 5.x index types (BTREE, TEXT, POINT, FULLTEXT, LOOKUP) based on query predicates, data cardinality, and consistency requirements. Includes a comparison table, trade-off analysis, and concrete validation steps with Cypher commands.
Variable-length patterns like MATCH (a)-[:KNOWS*1..3]->(b) enable powerful graph traversals, but can cause performance issues on dense graphs. Understanding the execution model helps you optimize before deployment.
Load CSV data into Neo4j 5.x safely: create constraints and indexes, run a verified import, and recover from partial loads without risking production data.
When a MATCH on a constrained property still scans nodes, the constraint may exist but the planner isn't using its automatic unique index. This guide shows how to verify constraint and index state, interpret PROFILE, and fix common schema and cluster causes.
Composite property indexes can collapse multi-attribute graph queries into a single index seek, but only when property order matches your query patterns. Here is how to decide when they help and what they cost.
Stop using slow CONTAINS queries in Neo4j. Learn how to implement full-text indexes, choose the right analyzer for your language, and validate search relevance using Lucene scoring.
Learn how to speed up Neo4j bulk imports by batching Cypher statements in explicit transactions, with a worked example and trade‑offs.
SKIP/LIMIT forces Neo4j to scan and discard every preceding row on deep pages. Keyset pagination with an indexed anchor and a compound cursor keeps page cost constant — here's how to choose, implement, and validate it.
Stop slowing down your Neo4j ingest with individual MERGE statements. Learn how to use UNWIND and parameterized batches to increase throughput and reduce network overhead.
neo4j