Data Storage: NoSQL Toasters and a Cloud of Kitchen Sinks
summary
Argues that the decisive choice is access pattern not buzzwords and gives a concrete framework called Access Patterns tying scheduled vs spontaneous and static vs dynamic to storage tradeoffs. Surveys real engines: mongodb, redis, cassandra, hbase, couchdb, marklogic, neo4j, react, couchbase, dynamodb and mentions Titan, exist db, atomic and gremlin/xquery. Drops surprising data: 90% of the data was created in the last two years (~2–2.5 zettabytes). Recommends converting dynamic queries into precomputed static views or hybrids (PostgreSQL metadata plus key value) and shows those approaches scale in practice.
AI-written summary · about
chapters (12)
- 0:00Start
- 5:38Mongodb
- 6:25Redis
- 6:42Cassandra
- 7:37Hbase
- 14:50Why no Sequel
- 16:27High Availability
- 23:35Why no Sequel
- 24:47Access Patterns
- 24:58Query Patterns
- 25:32Static versus Dynamic
- 26:30What Data Storage Solutions Are Appropriate for the Access Pattern
related talks
38mNot all data is created equal: NoSQL
45mNoSQL: Death to Relational Databases
38mNoSQL Databases: Its not a free lunch
57mMulti Tenant NoSQL and NewSQL Cloud data Design Patterns Deep dive
52mData storage for modern applications - Mike Lehan
01h 01mNoSQL: An Analysis
35mAn Ode to 17 Databases in 39 minutes
27mNavigating Unstructured Data - Availability vs Analytics in NoSQL