Cloud Object Storage
with an S3-compatible API

Yes. Because the API is S3-compatible, standard tools like the AWS CLI's s3 sync or rclone copy data straight across once you point them at your Krystal endpoint and keys. There's no special import step. It's the same commands you'd already use, aimed at a different endpoint.

Yes, and it's a common one. Any backup tool that uses S3, restic, the AWS CLI, a database dump piped to a bucket, can write here. You can send VM and database backups to a private bucket on a schedule. Keep those buckets private and reach them with access keys.

You authenticate programmatic access with access keys and keep buckets private by default. Make public only what needs to be. For example, the assets behind a static site. You leave everything else, like backups and exports, private.

For public assets that get heavy traffic, yes. A CDN serves cached copies from the edge, so requests don't all land on the bucket. This can keep response times down and reduces the download traffic billed against your Object Storage. For private data or low-traffic assets, serving straight from the bucket is fine.

You choose, with data centres in London, Amsterdam or New York. Point your tools at that region's endpoint and pick the one that suits your data-residency requirements or is closest to the VMs and users that read from it.

Both, as two separate meters. The £5/month includes 250GB of stored data and 1TB of transfer, and past those you're billed per gigabyte for what you use. It helps to know which one any of your workloads lean on. An archive is mostly storage and barely moves, while busy public assets are light on storage but heavy on transfer.