Object storage at a glance — S3, Blob, GCS, Spaces

Why this matters

Object storage is the workhorse of the cloud — it is where logs, backups, ML datasets, static assets, and user uploads land. The concept is identical across providers (buckets of immutable blobs addressed by key), but the defaults around encryption, versioning, public access, and consistency differ in ways that bite you on day two. Learning the four side by side is a once-and-done investment.

Demo

Create a bucket, upload a file, and list it on each cloud with the same handful of CLI commands.

Try it yourself

1. Create a bucket on at least two of the four clouds, upload hello.txt, and confirm the head-object/describe call shows server-side encryption is on by default.
2. Try to make the bucket publicly readable on AWS without disabling Block Public Access first — observe that the operation is refused, and read the error message carefully.
3. Compare the default storage classes (S3 Standard vs Azure Hot vs GCS Standard) and identify the cheapest cold-storage class on each — write down the price-per-GB-month for each.

Prompt your AI

Use these three in order. Each builds on the one before.

1. Basics & terminology

In one paragraph, explain object storage like I'm new to it, with a concrete S3 example of buckets, keys, and objects.

2. Why it works (the mechanism)

Walk me through how an S3 multipart upload actually works step by step — initiate, upload parts, complete, and how the ETag is computed differently than a single-PUT.

3. Advanced — application & what's next

Given a 2 PB ML training dataset that is read once a week and hot for 48 hours during each run, design the storage class + lifecycle policy on S3 vs GCS and justify the cost difference.

References

• Amazon S3 — User Guide
• Google Cloud Storage — Bucket overview