Offsite backup and disaster recovery in the EU
A second copy in another country, on other hardware, that a bad night at your primary site cannot reach.
Who it is for
A company whose primary site carries real risk
The backup drive is in the same building as the thing it is backing up.
A team with an RPO nobody has tested
There is a backup job and it reports success. No one has restored from it this year.
A team that has already been hit once
Ransomware encrypted the shares, and then reached the backup volume because it was mounted on the same host.
What you get to write to
- Storage priced per terabyte on RAID arrays in a Tier III facility
- FTP, FTPS, SFTP and SCP for whatever your current script speaks
- SAMBA/CIFS so Windows can mount it as a drive
- S3 for tools that only know object storage
- rsync, and rsync over SSH, for incremental sets
- HTTP and WebDAV through NextCloud when people need a folder rather than a protocol
- Daily backups of a cloud VM with retention set on a slider, up to 7 days
- Snapshots taken before any risky change, rolled back in two clicks
Sized by how much you need to keep offsite
The shared storage link runs up to 50 Mbit, so plan the restore window as well as the copy: 1 TB is roughly 45 hours at that rate. Keeping a hot copy on a cloud VM alongside the cold one is the usual answer.
What is and is not included
- Backup software licences such as Veeam or Acronis are not included. The storage is protocol-agnostic, so keep the tool you already run.
- We do not read your data and we cannot decrypt it for you. Encrypt before upload if it is regulated - you hold the keys.
- A disaster recovery runbook is not included. We provide the second site; who declares a failover, in what order, and how you tell customers, is yours to write.
- A restore test is not automatic. Book one - an untested backup is a hypothesis, not a copy.
RPO and RTO without the jargon
- RPO is how much data you accept losing. Back up once a night and your RPO is up to 24 hours, whatever the contract says.
- RTO is how long you accept being down. It is usually set by bandwidth and by who has the passwords, not by the backup itself.
- Three copies, two kinds of media, one offsite. The offsite one is what this page is about.
- Immutability beats cleverness: a copy production cannot delete survives ransomware, and a backup volume mounted on the production host does not.
- Prague and Covilha are different countries, different power grids and different upstreams under one AS - which is what makes them a real pair rather than two racks in a row.
One site versus two
| Backup in the same site | Second copy at DCXV | |
|---|---|---|
| Building-level incident | Both copies gone | Copy is in another country |
| Ransomware reaching the backup | Likely, if the volume is mounted | Separate credentials, separate network |
| Restore of a single file | Fast | Fast, over any of the protocols |
| Full restore speed | Limited by local disk | Limited by the link, so plan for it |
| Cost model | Hardware you buy and replace | Per terabyte, monthly |
| Proof for an auditor | A screenshot of a job | A second certified facility and a retention policy |
Standing it up
Decide what must survive
Not everything needs offsite. List the systems that would stop the business, and how old a copy of each you could live with.
Pick the protocol
Whatever your current tool already speaks. Storage is provisioned per terabyte and reachable the same day.
Seed the first full copy
The first upload is the slow one. Run it over a weekend, then keep it incremental.
Restore something on purpose
Pick a file, then a whole VM, and time it. That number is your real RTO - the rest is documentation.
Why choose us
- Tier III certified facilities, 99.982% facility SLA
- Own network, AS204057, IPv4 and IPv6 dual-stack
- 24/7/365 support with ~10 minute average response
- Cyprus company, EU jurisdiction, GDPR-native since 2007
FAQ
- How much does offsite backup cost?
Storage is priced per terabyte: 1 TB is EUR 14.41 per month, 5 TB is EUR 20.30 and 25 TB is EUR 49.80, which is the largest single volume we sell. Daily backups of a cloud server are configured on the server itself with retention of up to seven days
- What can I write backups with?
Whatever your current tool already speaks: FTP, FTPS, SFTP, SCP, SAMBA/CIFS, S3, rsync, rsync over SSH, and HTTP or WebDAV through NextCloud. The storage is protocol-agnostic on purpose, so Veeam, Acronis, Bacula, restic or a shell script all work without us installing anything
- How quickly can I restore?
A single file is immediate. A full restore is limited by bandwidth: the shared storage link runs up to 50 Mbit, so 1 TB takes roughly 45 hours. That number, not the backup job, is usually your real recovery time - which is why teams that care keep a hot copy on a cloud server as well as the cold copy
- Are the two locations genuinely independent?
Yes. Prague and Covilha are in different countries, on different power grids, with different upstream carriers, and around 2000 km apart. They share only our AS204057 routing. That is what makes them a real disaster recovery pair rather than two racks in one building
- How long are cloud server backups kept?
Retention is yours to set, from none up to seven days, and it can be changed at any time. Backups are separate from snapshots: a snapshot is a point you take deliberately before a risky change and roll back to in two clicks, while backups run on a schedule without you
- Can ransomware reach the backup copy?
Not if you do not hand it the keys. The offsite copy uses its own credentials on its own network, so an attacker who owns the production host does not automatically own it. What does get encrypted is a backup volume left permanently mounted on the machine it is protecting - that is the mistake worth designing out
If you require assistance or have additional questions, please contact the managers or write to the support team at support@dcxv.com
Ready to get started?
- Monthly billing
- No setup fee
- No lock-in