[AWS Cloud Fundamental Notes] Overview of Common AWS Architectures
Quickly understand common architectures for various applications.
Introducing various common application architecture components and their evolution.
Evolution of Architectures:
By narrowing the scope of development features, the deployment cycle of features is shortened.
To address the issue of insufficient system hardware: from physical construction to cloud classic architecture.
To accommodate more users: from cloud classic architecture to high concurrency architecture.
To reduce downtime: from high concurrency architecture to agile architecture.
Classic Cloud Architecture
| AWS Service | Function |
|---|---|
| Route 53 | DNS Service |
| EC2 | Servers |
| RDS | Database |
| S3 | File Storage |
| IAM | Permission Management |
| CloudTrail | User Behavior Monitoring |
| CloudWatch | Resource Monitoring |
Serverless Architecture Overview
Network packets enter the scheduling management service, which activates application servers to process packet requests.
- API Gateway: Scheduling management service
- Lambda: Application server
Each function is packaged separately and can be quickly started when needed, without affecting other functions.
Mobile Application Overview
Two major requirements for mobile apps:
1. Social Login
Authenticate identity through social software without the need for re-registration.
2. Message Notification
Send in-app notifications and emails to users.
High Concurrency Architecture
Traffic passes through CloudFront
to reach ELB
, then scales machines through Auto Scaling
.
Long-term tasks are forwarded to SQS
for processing by company servers, with statuses written to ElastiCache.
Relevant data is written to DynamoDB.
CloudFront
Amazon CloudFront is a content delivery network (CDN) service.
ELB
Distributes network traffic to improve application scalability, automatically distributing incoming application traffic across multiple targets and virtual devices in one or more availability zones (AZs).
Auto Scaling
AWS Auto Scaling monitors applications and automatically adjusts capacity.
SQS
A temporary storage for task messages, extracting them for processing when there are server resources available.
