- Fundamental cloud-compute model
- Client-server model
- A client makes a request, and the server responds to the request
- In AWS, a server would be a EC2 (Amazon Elastic Compute Cloud)
- Key concept to AWS
- Only pay for what you use
- Elastic means flexible; if you need more compute power, it can spin up. When you don't, it spins down.
Cloud Computing
- The on-demand delivery of IT resources over the Internet with pay-as-you-go pricing
- IT resources are fundamental, broadly similar tools used by different companies
- Over the internet means it can be served to any device with Internet connectivity
- Pay-as-you-go means you pay for resources based on need and use, not simply having them, available
- Offers flexibility of being able to request and release resources as needed, only paying for what's in use
- Significantly reduces CapEx by increasing OpEx
- Reduces the impact of under- or over-estimating required resources
- By sharing hardware resources, individuals and companies with reduced need can pay for less and have lower idle resources than self-hosted solutions
- Resiliency and global reach of cloud servers allows for rapid access across the world
Deployment Models
- Cloud-based
- All parts of the application are run in the cloud
- These applications could be designed and built for cloud use, or migrated from existing infrastructure
- On-prem (private cloud)
- Applications are deployed using on-prem hardware
- Applications can be run in virtualized environments or on dedicated servers
- Hybrid
- Cloud-based resources are connected to on-prem infrastructure
- This can be useful to meet compliance or regulations, while offloading non-regulated or other resources to the public cloud