Please enable JavaScript.

Coggle requires JavaScript to display documents.

Redshift - Coggle Diagram

- - - - enables fast query performance as the data warehouse grows.
  - - - similar data is stored sequentially on a disk.
    - - doesn’t require indexes or materialized views
    - - selects the appropriate compression scheme
  - - - takes advantage from MPP
- - - - provides monitoring using CloudWatch and metrics for compute utilization, storage utilization, and read/write traffic
    - - storing the backups
        
        user-defined retention period.
      - data durability
      - easily enabled to a second region for disaster recovery.
        
        only supports Single-AZ deployments and the nodes are available within the same AZ if the AZ supports Redshift clusters.
    - - Concurrency Scaling feature
        
        supports
        
        unlimited concurrent users
        
        concurrent queries
        
        automatically adds cluster capacity
  - - - based on PostgreSQL
    - - using
        
        ODBC
        
        C programming language interface that makes it possible for applications to access data from a variety of database management systems (DBMSs)
        
        JDBC connections
        
        Java API that manages connecting to a database
      - distributes & parallelize queries across multiple physical resources
    - - SQL based clients
      - business intelligence
- - - - with two or more compute nodes, an additional leader node coordinates the compute nodes and handles external communication.
      - dependens on dimension of workloads
  - - - distributes the compiled code to the compute nodes
      - assigns a portion of the data to each compute node.
    - - to the compute nodes only when a query references tables that are stored on the compute nodes
        
        All other queries run exclusively on the leader node.
  - - - Node Slices
        
        allocated portion of workload assigned to the node
        
        by leader node
        
        numb. of slices determinated by node size of cluster
        
        distribution key
        
        column of a table
        
        parallel processing to load data
        
        execute queries efficiently.
    - - direct access not allowed
    - - send intermediate results back to the leader node for final aggregation
    - - dedicated CPU
      - memory
      - attached disk storage
      - 1-128 nodes
    - - Dense Storage
        
        creation of very large data warehouse
        
        usind HDDs
      - Dense Compute
        
        very high performance
        
        using fast CPU
        
        large amount RAM
        
        SSDs
  - - - where
        
        Data warehouse data is stored in a separate storage tier
      - ability to scale the storage to petabytes
        
        using S3 storage.
        
        advantage as
        
        data block temperature
        
        data block age
        
        workload patterns
      - enables scale and pay for compute and storage independently
        
        cluster can be sized based only on the compute needs
      - tier-1 cache.
        
        high-performance SSD-based local storage