Cloud

Not slower. Our datacenters in Madrid, Munich, Helsinki, and Amsterdam run the same hardware generations as the hyperscalers. Latency within Europe is lower when your users and your infrastructure are on the same continent, the one whose law you comply with. On price, the cost of a data breach, a regulatory fine, or a forced migration later tends to be higher than any difference today.

On a VPS, CPU and RAM are virtualised and shared with other customers on the same physical host. A VDS runs on a separate cluster where the resources are yours alone, not shared. No noisy neighbours. For critical production we recommend VDS. Two more variants sit on that same footing: MEM, memory-optimised, and MNG, which we also manage for you.

Everything runs on KVM. That means standard virtio drivers, no proprietary guest agent, and images that behave the same way they do anywhere else you have run KVM. If you are moving an existing disk in, talk to us before the cutover and we will size the target with you.

Every instance is dual stack, IPv4 and IPv6, from the moment it is created, and there is no separate order for it. IPv6 ships as a single /128, and a /56 is rentable if you need to subnet. Additional IPv4 is rentable the same way, in blocks from a /29 up to a /24.

In our own European datacenters in Madrid, Munich, Helsinki and Amsterdam, all four live. You pick the site when you order, and your data does not leave the European Union.

Yes, and you do not have to reach them over the open internet. We peer directly with AWS, Azure, GCP, Oracle and Cloudflare at European exchange points, so hybrid traffic takes a short, known path instead of a transit provider's.

You do, on VPS, VDS and MEM. Those three are unmanaged: we run the platform and the network, you run everything from the OS upwards. MNG is the managed type, and there the patching, the monitoring and the backups are ours. MNG is also the substrate the managed products stand on, so managed databases and managed Active Directory run on it rather than beside it.

Usually you do not find out. An instance is not tied to one physical machine, so when a node fails the workload moves to another one. That is the difference between this and a dedicated server, where a failed component is a ticket and a repair window. It is also why the availability we commit to is a platform number rather than a per-machine one.

With instance snapshots and full backups, both features of the platform itself. Object Storage is a separate product you may also want, for assets and archive, but it is not the backup answer: you should not have to buy a second product to get something the instance already does.

Support is free on every product we sell, with a 4-hour target response time on support requests. The advanced plan is paid and buys shorter targets plus weekend and out-of-hours cover.

99.9% product availability, by contract, with automatic credits if we miss it, and the same figure covers every product we tie to an SLA. On a virtualised platform that number means what you would want it to mean: because a failed node moves your workload rather than stopping it, what is being measured is the service rather than a particular machine.

Standard instances, whatever the type, are delivered the same working day. Custom configurations, with dedicated hardware or specific networking, are confirmed with a firm date in the quote.

Each instance is either on a monthly bandwidth allowance or on unlimited traffic, and you pick which when you order. If you take the allowance, traffic above it is billed per additional block, or blocked if you ask for that when you order. Either way it is in your contract, not a surprise line on an invoice.

Yes, and we have done it many times. Our architecture team runs honest migration assessments: we tell you what is straightforward, what is complex, and what is probably not worth moving. We also peer directly with AWS, Azure, GCP, and Oracle at internet exchange points, so migration traffic takes a short, known path over direct peering rather than whatever a transit provider chooses. That makes transfers faster, more predictable, and cheaper than most customers expect.

One month on the instance itself. A savings plan is a separate commitment and it behaves like one: take 12 or 36 months for the lower rate and you pay that plan out for its term, whether or not you keep the instance running. That is the trade, and it is yours to make. Without a plan, the minimum stays one month.

We delete the disk and zero the space before it goes to anyone else, and we do the same with your snapshots and backups. What we do not offer for cloud is the NIST SP 800-88 erasure we carry out for bare metal, and the reason is what that standard is written for: a named physical drive, sanitised while it is out of service. Bare metal gives us that moment, because the machine comes out of service when you leave. Cloud storage stays in service, so there is no drive to name. Where the large clouds cite the standard, they mean a drive at the end of its own working life, not the moment you press delete. If your compliance regime asks for that erasure, it needs bare metal.

GDPR requires that personal data is processed under a legal framework that guarantees its protection. Storing data with a US-owned provider creates a structural tension: EU data protection law on one side, US extraterritorial reach on the other. Running on DIVERSITY removes that tension. Your data stays in EU datacenters, under EU jurisdiction, with no foreign parent company that could be compelled to act against your interests. For regulated industries, that is not a nice-to-have. It is the baseline.

It means the company running your infrastructure is incorporated, operated, and legally subject to European law. Not a subsidiary answering to a US parent. When a government or regulator issues a demand for your data, the answer is determined by EU law, not a foreign jurisdiction. That is not a feature we added. It is how we are built.

The US CLOUD Act lets American authorities compel any provider under US jurisdiction to hand over data in its possession, custody or control, wherever it is stored, without notifying you. If your provider is ultimately owned by a US company, the Act reaches your data through that parent. DIVERSITY has no US parent: no US company owns or controls us. There is no entity above us that can receive such an order.

Encryption protects data in transit and at rest from third parties. It does not protect against orders served on your cloud provider under their home jurisdiction's law. If the provider holds the keys, or is compelled to assist in accessing your data, encryption alone is not a legal shield. Jurisdiction is.