A second site, planned, built and tested with you.

Then both go together: a second site of ours protects you against losing a site, not against losing us. If that is the risk you are planning for, keep your main systems where they are, on AWS, Azure, Google Cloud or in your own datacenter, and recover to us. Or run with us and keep your recovery site in your own datacenter, over Private Link. Either way, addresses of your own stay yours, and we can announce them with your own AS.

The latest copy of the virtual machines you choose, ready to start, refreshed on the schedule you agree. Alongside it, as your plan sets: a replica of your managed databases, ready to take over; a copy of your backups, which can be locked; and capacity set aside, if your contract includes it. The ready copy is the latest one; your backups reach back as far as their schedule keeps them. The plan decides which systems get a ready copy and which rely on backups.

Through a process of our own. On the schedule you agree, it clones each machine you choose, extracts its data and imports it into our datacenter, so the latest copy is ready to start there. Your VMware estate stays where it is; if it runs in your own datacenter, it reaches ours over Private Link.

Yes, and that makes it a second provider rather than a second region: another region of the same provider is still the same company, under the same contract and the same law. The copies are made with our own tooling or with the tools you already use, on the schedule you agree. We peer directly with AWS, Azure and Google Cloud at European exchange points, so your copies take a short, known path to us instead of a transit provider's. A service that only your provider runs has no copy to start here: the plan names those systems.

Yes, in either direction: your datacenter as the main site and ours for recovery, or the reverse, depending on where most of your systems run. It connects to ours only over Private Link: through our router, over our fibre, or through a tunnel over your own connection.

Your own addresses, announced from two of our sites, so your services keep their addresses after a failover. We can announce them from our AS, run BGP with your own AS, or both. With your own AS, you can also announce them through another provider alongside us. If you have no AS number or address space yet, we help you apply for them. Routes take time to settle after a change, so it is not instant. For services without addresses of their own, health-checked DNS answers with whichever site is up; that is not instant either, because users move over as the answers their resolvers have cached expire.

Yes, within this setup. An unmanaged dedicated server has no backup of its own, and no dedicated server has failover underneath it; this setup adds backups of it in a second site, as your plan sets. Our managed databases do not fail over between sites on their own: a replica in a second site is something this setup adds.

It can be, or it can wait for your decision: the design and your contract set which, system by system, and the runbook names who decides. Who then carries out the steps, our engineers or your team, is set in your contract too.

Three things: we watch replication and tell you when a copy falls behind, we keep the runbook current as your systems change, and we run the recovery tests with you.

We restore from a backup taken before the attack, if your backups reach back that far, into an isolated network first. We check it with you, and only then bring it back. Responding to the attack itself belongs to our SecOps service.

On a schedule you choose. Your production keeps running in the main site while we start the copies in the second site, check them with you and shut them down. Each test is run from the runbook and ends in a written report of what came back and what changed.

As one setup designed for you. The contract names the systems, their objectives, where each copy waits, who decides to fail over and who carries it out, and the recovery tests, on a schedule you choose. We do not publish prices.

Yes, if your contract includes it: capacity set aside for you in the second site.

With the discovery call, and it is free: 60 minutes, an engineer rather than a salesperson, and a written diagnosis inside a week. It is a conversation, not a plan, a test or a design. Already a customer? Same call, same form.

No. Neither requires your data, or your recovery site, to be in the EU. Both make you answer for your continuity and for your providers, and your contract's DORA addendum gives you and your supervisor the audit, access and inspection rights that come with that. European supervisors now publish a list of the critical ICT providers the financial sector leans on hardest, and the large hyperscalers are on it. We are not, so a recovery site with us does not add to that concentration.

The copies we keep sit in our datacenters, in the site you choose: Madrid, Munich, Helsinki or Amsterdam, all four in the EU. Your evidence says which country each one is in. If your own datacenter is the recovery site, or copies are kept anywhere else, by your own tools or another provider, those copies are outside that answer.

For the copies we keep: DIVERSITY is a European company with no parent company, US or otherwise, so there is no entity above us for an order from outside the EU to be served on. That is a structural answer rather than a policy one.