The NoSQL movement @ DotNetToscana
Download as PPTX, PDF1 like702 views
The document discusses the pros and cons of relational databases (RDBMS) versus NoSQL databases for modern web application requirements. It outlines some key characteristics of RDBMS like SQL queries and transactions, and notes they may not be the best fit for requirements like large heterogeneous datasets, high user volumes, and dynamic needs. The document then categorizes different NoSQL database types and provides examples for each, concluding that a polyglot or multi-model approach may be most suitable.
The NoSQL movement @ DotNetToscana
- 1. Matteo Baglini www.dotnettoscana.org Software Developer, Freelance [email protected] http://it.linkedin.com/in/matteobaglini http://github.cpom/bmatte
- 2. No more SQL, awesome!!! 2
- 3. Yes but, why? 3
- 5. High number of heterogeneous content 5
- 6. Many connected users 6
- 7. Immediately push info to the users 7
- 8. Collection of large data sets 8
- 9. Scalable and on-demand architecture 9
- 10. Dynamic requirements 10
- 11. RDBMS 11
- 12. Tables and relations Schema-full Pre-defined structure Transaction (even Distributed) Consistency Declarative query language (SQL) 12
- 13. Is it the right tool? 13
- 14. Alternative Database (NoSql) 14
- 15. Alternative structures Schema-free Non transactional Eventual consistency Programmatically query routine Easy to scale-out Use case oriented 15
- 16. NoSql Categories 16
- 17. Data (Value) was identified by a unique Key The Value can be anything Primarly queriable by Key Examples: Amazon SimpleDB, Azure Table Storage, Riak, Redis, Voldemort, MemcacheDB. 17
- 18. Data contained into Documents that was identified by a unique Key Document was stored as JSON object (properties and values) Values can be scalars, arrays or complex types Queriable by Key or MapReduce Examples: MongoDB, CouchDB, RavenDB 18
- 19. Data contained into Tables and Column Families Each Column Families was a set of key-value pairs Partial schema Queriable by Key, MapReduce or custom langs Examples: Google BigTable, Hbase, Cassandra 19
- 20. Data contained into Nodes and Edges Each Node was a set of key- value pairs Each Edge represent a relationship Queriable by custom langs Examples: Neo4j, OrientDB, Titan, Sones 20
- 21. NoSql or RDBMS ? 21
- 22. Polyglot Persistence 22