Please enable JavaScript.

Coggle requires JavaScript to display documents.

System Design 2 - Coggle Diagram

- - - - Store data in Tables where each row represent an entity and each column represent data about that entity.
    - - Different Storage Models: Key-Value, Document, Graph and Columnar.
  - - - Dynamic: columns can be added on the fly and each "row" (or equivalent) doesn't have to contain data for each column.
    - - Fixed: columns must be decided and chosen before use. The schema can be altered but involves modifying the whole db.
  - - - Use of Structured Query Language which is very powerful.
    - - Unstructured Query Language UnQL, usually focused on collections of documents.
  - - - Vertical (Horizontal is also possible but is challenging and time consuming)
    - - Horizontal
  - - - SQL
        
        Yes
      - NoSQL
        
        No: most of the NoSQL solutions sacrifice ACID compliance for performance and scalability.
- - - - Voldemort
      - Dynamo
      - Redis
    - - Store data in documents grouped in collections
        
        CouchDB
        
        MongoDB
    - - Instead of Tables define columns families to represent entities.
        
        Cassandra
        
        HBase
    - - Neo4J
      - InfiniteGraph
- - - - Client performs a request and waits for the server to respond.
      - Problem
        
        Client has to keep asking the server for any new data. A lot of responses are empty, generating an HTTP overhead.
      - If no new information is available, an empty response is returned.
    - - Client performs a request and waits for the server to respond.
      - If no new information is available, the server holds the request until information is available.
- - - - System continues to function even if the communication fails between nodes
    - - Every request gets a response even with node failures.
    - - All nodes see the same data at the same time.