4 types of databases in nosql:
- document store
- document is a unit that is flexible
- wide column database
- key-value store
- graph database
| document store | wide column database | key-value store | graph database | |
|---|---|---|---|---|
| store data in semi-structured records, using formats like json, bson or xml ⇒ highly flexible, allow store data in a format similar to application’s code objects | extreme scalability, efficient data compression, ideal for massive datasets | every data item is stored as a unique key paired with a corresponding vlaue ⇒ optimized for lightning-fast read and write operations but are limited in terms of complex querying | focuses on relationships btw data points by using “nodes” (entities) and “edges” (relationships) ⇒ for traverse complex, highly connected data much faster than traditional sql joins | |
| usage | content management systems (CMS), ecommerce catalogs, user profiles | large-scale data warehousing, real-time analysis, iot sensor data | caching, session management, real-time leaderboards | social networks, fraud detection, recommendation engines |
| exmaples | mongodb, apache couchdb, firebase firestore | redis, amazon dynamodb, memcached | neo4j, amazon neptune, arangodb |
mongodb:
- basc unit data: ddocument
- collection : containers for documents
- schema flexible -
- mongosh
when is a nosql >> relational db?
- when handling unstructured, dynamic data
- massive scalability needs
- rapid devlopment cycles
- simpe data retrieval
- large-scale performance