High Availability, Scalability, ELB, ASG

info

These were the topics I created flashcards for (Remnote) and would revise them using spaced repetition. The formatting is an export from Remnote.

High Availability & Scalability
- Why are distributed systems the basis for scalability being horizontal or vertical?
  - data writes ‒ differentiation
    - integrity = how do you ensure the data you're reading is accurate? i.e. the most up to date.
    - durability = confident our "writes" are committed to the database before continuing.
  - compute, capacity, data reads ‒ commodity
    - if its generic, the same unit type, "undifferentiated" it can be scaled
- Active High-Availability vs Passive High-Availability
  - Active-active ~ 2 x call centre buildings with operators working at both at the same time.
  - Active-passive ~ 2 x call centre buildings with operators, 1 x is actively taking calls, the other is on standby in case building one goes down.
Elastic Load Balancing
- single-point of contact, distribute seamlessly across downstream instances.
- Types
  - Classic Load Balancer
  - Application Load Balancer
    - operates at the Application Layer (Layer 7)
    - "target groups" are {{downstream}} {{logical}} groups to send traffic to.
    - health checks happens at the {{TARGET}} GROUP level.
    - target groups for ALB can be― ↓
      - IP addresses (must be private)
      - EC2 Instances
      - Lambda Function
      - ECS Task
    - protocols supported ↓
      - HTTP
      - HTTPS
      - HTTP/2
      - Websocket
  - Network Load Balancer
    - operates at Network Layer (layer 4)
    - protocols: TCP, UDP.
    - if you want "Fast'!
  - Gateway Load Balancer
- Stick Sessions
- Cross-Zone Load Balancing
- SSL Certificates
- Connection Draining
- Auto Scaling Groups
  - Scaling Policies