NoSQL describes storage that does not rely primarily on SQL or standard relational tables and relationships. It does not mean that a database cannot use SQL. NoSQL has no shared database type or standard, and systems within the same category can work differently. section 3.6
Choose a model to suit the workload:
- Document stores, such as MongoDB and Couchbase, suit frequently retrieved complex data when entity relationships matter less. section 3.6
- Key-value stores, such as Redis and DynamoDB, suit distributed caches where fast retrieval matters most. section 3.6
- Graph databases, such as Neo4j and ArangoDB, represent connections and suit maps of relationships such as social networks or supply chains. section 3.6
- Column-family stores, such as Cassandra and HBase, allow columns to vary between rows and suit data warehousing. section 3.6
In Aspire, popular NoSQL databases and Azure-native storage services have integrations. Some Azure services can use emulators for local development. section 3.8
