S3 object storage with a fixed quota, in the EU

S3 is the interface Amazon introduced for storing files over HTTP, and it became the standard the whole industry speaks. Compatible means your tools do not care who runs the storage: you enter an endpoint address, an access key and a secret, and Veeam, rclone, Cyberduck, a Synology NAS or your own application code work exactly as they would with Amazon. Files are stored as objects inside buckets rather than in folders on a mounted disk, which is what lets the storage scale independently of any server and stay available when that server is not.

Most storage clouds meter what you consume and bill it afterwards, which means the invoice is only known at the end of the month and a mistake in a backup script becomes a financial event. We sell a reserved quota instead: you pick the size, that capacity is held for you, and the price never moves. When the quota is full, further writes are refused until you delete something or move to a larger plan. It is a deliberate trade: you lose elastic growth and you gain a bill you can predict.

No. Downloads are included, and there are no per-request charges either. That matters most in the situation storage exists for: restoring a large backup. With providers that bill egress, a full restore can cost more than months of storage, which quietly discourages testing whether backups actually work. Here a restore is just a restore. We do ask for common sense rather than using the storage as a public content delivery network.

In the Netherlands, replicated across three independent Tier 3 data centres, so the loss of one building does not take your data with it. Everything stays under EU jurisdiction, which keeps GDPR questions short: your files are processed inside the European Union by an EU company, and you can say so in your own documentation. If you also need a copy in Latvia, our Riga data centre can host the servers that write to this storage.

Yes, Veeam is the tool this storage is most often used with. You add it as an S3-compatible object storage repository, paste the endpoint and keys, and back up directly to it. Object Lock is supported, so you can make backups immutable for a chosen retention period and satisfy the requirement that ransomware must not be able to destroy your copies. Proxmox Backup Server, Bacula, Restic, Duplicati, MSP360 and Synology Hyper Backup work the same way.

Object Lock marks an object as unchangeable until a date you set. Until then nobody can overwrite or delete it, not an administrator, not a stolen access key, not the ransomware that reached your backup server. This is the single most important feature in modern backup design, because attackers now go after backups first, and normal backups can be deleted by anyone holding valid credentials. Turn it on for your backup buckets and set a retention that matches how far back you would ever restore.

The everyday surface is there: buckets, multipart upload for large files, versioning, Object Lock, access control on objects and object tagging. Some newer or more exotic parts of the S3 API are not implemented, for example bucket policies, CORS configuration and full lifecycle rule sets, so if your application depends on one of those, ask us before ordering and we will check it against the platform rather than guess. Tools we have verified are the AWS CLI, rclone, Cyberduck and Veeam.

Within one business day. Each storage project is created by hand so that every order is reviewed, which is also how we keep the platform clean of abuse. You receive an email when the service is active, and the endpoint, access key and secret appear in your client area together with copy-paste configuration for the common tools. If you need it faster, write to support before ordering and we will tell you honestly when we can have it ready.

Yes. Each order creates its own project with its own endpoint and keys, so ordering twice gives you two fully separate storages, for example one for production backups and one for a client or a test environment. Inside a single project you can create as many buckets as you like, and they share that project quota. If you want separate quotas or separate credentials for different teams, order separate services.

Access is suspended seven days after the due date: the keys stop working, but your data is untouched and everything resumes as soon as the invoice is paid. If the service is still unpaid thirty days after expiry, the storage project and its contents are deleted permanently, and we send warnings before that point. This is the same logic as any other service with capacity reserved for you, and it is why the dates are stated here rather than buried in the terms.

Yes, larger capacity is available on request and is quoted individually, because at that size the price depends on the commitment and on how the data will be used. Write to us with your target size, how much of it you expect to write and read each month, and whether you need immutability, and you will get a straight answer including whether we are actually the right supplier for that workload.

The smallest plan is one month at 29 EUR, which is the honest way to try: order 1 TB for a month, run a real backup and a real restore, and see how it behaves with your tools. If it does not suit you, cancel before the next month. We would rather you test with a real workload than promise a free trial that proves nothing about how your own data behaves.