High Level Notes for the AWS Cloud Practitioner Certification

💻 Compute = (Brains CPU) & Memory (RAM)
🗄 Storage = disk or cloud storage (location where your data is saved)
🗂 Database = app that stores data in structured way that can be queried
🌐 Networks = home network (device that allows house to connect to internet, modem/router gives connectivity to internet)
Devices talk to each other and the internet, in order to talk to each other there is a switch (often times the router), allows communication to flow to the internet
Computer connects to switch, which connects to internet, which connects to DNS, looks up ‘facebook’ in its address contacts, and sends you to the right location for facebook
All of these different computers connected to the internet, within this you have all of these networks with computer and memory
Server sprawl = when servers had a lot of unused compute and or storage, so thats when the “cloud” started selling that extra capacity
Remote workers and corporate offices can all access data centers in the “cloud”

What are clouds made of?

SaaS = infra, os, and software provided
IaaS = part or all of an infra provided
PaaS = infra and os provided
stack = set of components grouped together for application
hypervisor = create virtual hardware (components that make up compute storage and network) allow you to create virtual machines (PaaS). Abstracts the infra away and can just move to new hardware for example.
private cloud = on prem, fully customizable, high cost and overhead, inelastic scaling, latency
public cloud = pay as you go, cheaper, low overhead, elastic scaling, some restrictions, faster/simpler deployments

Introduction to Cloud Terminologies

Scalable = can add additional capacity to meet compute and storage needs
Elasticity = grow and shrink according to demand
Fault tolerance = can withstand some failure and remain functional, can return to full capacity in wake of errors — i.e. two ports on the router, when 1 port fails use the other
Highly available = being able to access when you intend to — i.e. have multiple routers in case one is too fail, the second router is still available (not just 1 single point of failure)

AWS Console = UI interface with AWS services
API Access = can use terminal to access AWS services programmatically
Flexibility = able to choose as many AWS services as you need
Cost Effectiveness — Pay only amount of services you need, no long term contracts or up front commitments.

Every region has at least 2 availability zones (AZs)
Inside each AZ is a data center (1 or more)
All resources you provision go inside that data center within an AZ belonging to a Region.
Each AZ has high availability and fault tolerance with low latency between AZs
Networks (cables along ocean floor in some instances) connect regions together

Networking = setup own network to access any other resources on AWS
AWS Virtual Private Cloud (VPC) = basis for network infrastructure, can provision your own aws environment
AWS Direct Connect — can connect your corp env with a low latency connection direct to AWS
Route 53 — setup domain name services (DNS)
DNS — need to tell computer where mywebsite.com is in form of ip address
EC2 — virtual server (instance) of compute
Lambda — serverless compute, don’t need to worry about the config of the compute yourself
S3 — data store, bulk upload files for storage
Glacier — long term storage (dont need frequently but need to keep for regulatory reasons). Cheaper than normal S3 for infrequently accessed data.