S3 stores files (called objects) in containers (called buckets). It is infinitely scalable, globally accessible, and 11 nines (99.999999999%) durable. Think of S3 as a hard drive in the cloud with unlimited capacity.
| Storage Class | Retrieval Speed | Cost | Best For |
|---|---|---|---|
| S3 Standard | Milliseconds | $0.023/GB | Frequently accessed: websites, app data |
| S3 Intelligent-Tiering | Milliseconds | Auto-optimised | Unknown access patterns — auto-moves objects |
| S3 Standard-IA | Milliseconds (+ retrieval fee) | $0.0125/GB | Backups, DR — accessed monthly |
| S3 One Zone-IA | Milliseconds (+ retrieval fee) | $0.01/GB | Reproducible data — lower redundancy |
| S3 Glacier Instant | Milliseconds | $0.004/GB | Archives accessed quarterly |
| S3 Glacier Flexible | 1-12 hours | $0.0036/GB | Archives accessed 1-2 times per year |
| S3 Glacier Deep Archive | 12-48 hours | $0.00099/GB | 7-10 year compliance archives — cheapest! |
# S3 Use Cases
Static Website Hosting -- HTML, CSS, JS served directly from S3
Data Lake -- Store Parquet files for Athena queries
Backup and DR -- Automated daily backups from RDS, EC2
Content Distribution -- Origin for CloudFront CDN
Event-Driven Processing -- S3 upload triggers Lambda function
# S3 Performance
Per prefix: 3,500 PUT/s + 5,500 GET/s
For high throughput: use multiple key prefixes
For large files: use Multipart Upload (>100MB)