Emergence of NoSQL Data Stores

Emergence of NoSQL Data Stores

1. Relational Database

   • Not good at handling big data

2. Relational and OOP databases don't work well together

3. OOP Basics

   • Class
     • Object
   • Encapsulation, Polymorphism and Inheritance

4. Object Relational Impedance Mismatch

5. ORM (Object relational Mapping)

6. Enterprise Architecture

   • Integration Approach
   • Application Database

7. Scalability

8. NoSQL Technologies

   • neo4J
   • CouchDB
   • MongoDB
   • Cassandra

9. Why use NoSQL

   • Realtime interaction with users
   • Agility, Consistency, performance, scalability
   • Open Source

10. aggregate vs aggregate ingnorant

    • Relational Models are aggregate ignorant
      • relational for same data requires many tables
    • NoSQL sees aggregate as clusters
      • collection of data related to a unit

11. Graph database

    • neo4J
      • Nodes form network
        • Social networking connecting friends

12. Distribution models

    • High read
    • High write

13. Replication and Sharding

    • Sharding
      • Different data across different nodes
    • Replication
      • same data across different nodes

14. Replication & Master

    • Traffic only through master for write
      • inappropriate for high write

15. Replication Peer to peer

    • Traffic written to any node
    • greater resilience

16. Combining Master/Slave and Sharding

    • All write through the server
      • Different nodes have different data
        • If master fails must know which node to write too

17. Consistency Locks

    • Pessimistic
      • Lock table for editing at beginning of workflow
    • Optimistic
      • Verify with server before making changes at end of workflow