S3
Durable object storage for files, media, backups and data lakes.
Problem it solves: Databases are a poor place for large binary data.
Where it sits: Written by upload services; read via CDN.
When to use
- Media, files, data lake, model artefacts.
When not to use
- Small, frequently updated records.
Failure modes
- Per-prefix request limits.
- High first-byte latency without a CDN.
Metrics to watch
- GET/PUT rate
- Egress (Gbps)
- First-byte latency
Alternatives
- Distributed file system
Configuration in Labs (illustrative defaults)
- Capacity per instance (req/s) = 50000 — Maximum requests one instance completes per second before it saturates.
- Write capacity (primary) (writes/s) = 10000 — Writes the single primary can absorb per shard. Replicas do not add write capacity.
- Base service time (ms) = 40 — Processing time per request at zero load. Queueing adds to this as utilisation rises.
Vendor examples (pattern only): AWS managed service. Simulated as a generic “Object / blob storage”.