Enhance your work with graph database
summary
Company with 15M users and 14M catches moves a friends-of-friends email feature from slow SQL joins (up to 15s) and hourly caching to a Neo4j graph model with friends-of-friends query that runs in milliseconds and removes stale caches. Spam rings surface as dense mutual-connection clusters in graph queries. PageRank shifts from NetworkX (memory-bound, hours on partial data) to Neo4j + GDS, running on full data in ~30 minutes. Examples use Cypher, Bolt driver, and concrete data and performance gains.
AI-written summary · about
chapters (14)
- 0:00Introduction and welcome
- 1:51About the speaker and Fishbrain
- 4:58Why graph databases matter
- 8:50What are graphs
- 10:59Graph vs relational databases
- 13:27When to use graph databases
- 14:09Example 1: Transactional use case
- 21:19Example 2: Analytical use case
- 24:54Example 3: Data science use case
- 34:30How to start with graph databases
- 36:37Data modeling in graphs
- 39:51Resources and tools for graphs
- 43:00Conclusion and final thoughts
- 43:42Q&A session
related talks
01h 29mSpring Data Neo4J
47mWhat Business Can Learn from Dating
26mHow NEO4J Saved my Relationship
26mGraph Databases will change your (freakin') life
28mNeo4j for Beginners: Pros and Cons, Based on Social Networks
30mPerformant GraphRAG Improving Neo4j to Drive Autonomous AI ...
55mLet Me Graph That For You
49mModern Graph Databases