Run by us. Hosted where you choose.

Fully managed databases, in your own infrastructure or ours. Fully GDPR compliant and secure by default.

HOSTING

Your infrastructure or ours

Inside your own environment, or fully hosted by us at Hetzner in Germany or Helsinki, Finland.

ISOLATION

Your own environment

Nothing is shared with other customers.

SUPPORT

8x5, 16x5 or 24x7

On-call hours and response times follow your SLA. One SLA covers all your Cloud Exit services.

NO LOCK-IN

Standard open-source software

Your own team or another provider can take over the setup later.

Service

What we take off your plate

Your applications depend on servers, databases, storage, backups and security. Someone has to set them up, keep them updated, watch them and fix them when something goes wrong. We do that, so your team can spend its time on your product.

  • Databases

    PostgreSQL, MySQL and Redis, set up, kept up to date and copied to a standby, so a single failure does not take your application down. In your infrastructure or fully hosted by us, your choice.

  • Analytics data

    ClickHouse for large volumes of reporting, log and event data. We run it, back it up and upgrade it. Managed ClickHouse

  • Storage

    High-speed NVMe storage only, encrypted at rest.

  • Backups

    Encrypted, stored away from your systems and test-restored on a schedule.

  • Security

    Access limited to the people who need it, data encrypted, security updates applied on a fixed schedule, and audit logging available.

  • Managed secrets

    Passwords, API keys and certificates kept in one secure vault, changed regularly, with a log of every use. Managed secrets

  • Supporting services

    File storage, message queues, search and certificates, run to the same standard.

Offline environments

For systems that must never be connected to the internet. Software and updates come in through a controlled route, and every access is logged.

Managed databases, where you want them

PostgreSQL, MySQL, Redis and ClickHouse, run by us either way. You choose where they live.

OPTION 1

In your infrastructure

The databases run inside your own environment, on servers and accounts you control. We set them up and run them day to day.

Servers
Yours
Data
Stays in your environment
We handle
Setup, updates, backups, failover and monitoring

OPTION 2

Fully managed by us

We host and run everything end to end on our infrastructure in the EU. Your team connects to the database and uses it; the rest is our job.

Servers
Ours, in Germany or Finland
Data
In the EU, in a database only you use
We handle
Everything, from hardware to backups

Reliability

Reliable, Private and Secure

Reliable infrastructure comes down to a few habits, applied every time: few ways in, no change without a record, backups that are proven to work, and people who respond.

Only yours

Every customer gets a separate environment. Nothing is shared with other customers, so their problems cannot become yours.

No way in from outside

Your environment collects approved changes from us. There is no connection from our side into it, one less door to try.

Every change on record

Changes are written down, reviewed and recorded before they go live. You get a full history, and any change can be undone.

Backups that are tested

A backup only counts if it restores. We test restores on a schedule and keep every result on record for your auditors.

Updates on a schedule

Security updates go out on a fixed schedule for every environment we run, so none of your systems ever falls behind.

Engineers who respond

Our engineers are on call for everything below your application, with hours and response times set by your SLA.

Ownership

No Vendor Lock, EU-Native hosting

Nothing to get locked into

Everything we build uses widely used open-source software, with Kubernetes at its core, and the complete configuration is kept in version-controlled repositories. In your own infrastructure, the accounts and data are in your name, and your team can take over the same environment without moving anything. When we host it, the same setup can run in your infrastructure or with another provider.

Under EU law

When we host it, your environment runs in Hetzner data centres in Germany or Helsinki, Finland, operated by a European company under EU law. Your data stays in the EU, and no US cloud provider is involved.

Technical details for your engineersPlatform, delivery, isolation and security
Platform
Kubernetes, managed with Syself, on Hetzner compute and storage
Delivery
GitOps from a central management plane. Clusters pull their desired state, with no inbound access.
Isolation
A dedicated cluster per customer. Air-gapped clusters have no egress and use mirrored registries.
Data services
PostgreSQL, MySQL, Redis and ClickHouse with replication, failover and connection pooling. ClickHouse runs replicated and sharded with ClickHouse Keeper.
Security
Least-privilege identity, network policy between workloads, encryption in transit and at rest, image provenance, audit logging available
Secrets
OpenBao vault with rotation, scoped access and audit logging, synced into clusters with External Secrets Operator
Backups
Encrypted and off-cluster, restore-tested on a cadence, with results recorded as audit evidence
Storage
NVMe only, encrypted at rest

Pricing

Managed database pricing

The prices below are for the fully managed option, where we host and run everything. If we run the databases in your infrastructure, you pay a management fee per core plus a support SLA. See how that works.

All prices exclude VAT and are monthly unless stated otherwise.

FULLY MANAGED BY US

MySQL

Starting from€319per month

3 nodes with 2 vCPU, 8 GiB RAM and 100 GB NVMe each

€199 one-time setup

PostgreSQL

Starting from€319per month

3 nodes with 2 vCPU, 8 GiB RAM and 100 GB NVMe each

€199 one-time setup

ClickHouse

Starting from€216per month

1 shard with 2 replicas, 2 vCPU, 8 GiB RAM and 100 GB NVMe each. ClickHouse Keeper included at no extra charge

€299 one-time setup

Every database includes

  • Dedicated cores, no oversubscription
  • Datacenter NVMe
  • Monitoring
  • Alerting
  • Configured automatic failover
  • Security patches
  • Minor upgrades
  • Backup scheduling
  • Restore checks

The price is the resources above (compute, memory, NVMe storage and backups) with no management fee on top.

Without an SLA: we monitor your databases around the clock and fix platform incidents as soon as we detect them. You reach us by email at contact@cloud-exit.com, or optionally through a shared Slack channel, at no extra charge during business hours (09:00 to 17:00 Amsterdam time (CET/CEST)). That support is best effort, with no guaranteed response times. An SLA gives you hard guarantees: contractual response times and on-call hours, 8x5, 16x5 or 24x7. One SLA covers all your Cloud Exit services.

IN YOUR INFRASTRUCTURE

Per core, plus your support SLA

One SLA covers all your Cloud Exit services

Management
€15 per core per month, for every core we run
Support SLA, required
8x5 €2,500 · 16x5 €4,250 · 24x7 €12,500 per month. 8x5 is the default.
Not included
Your servers, storage and any third-party platform fees, such as Syself Autopilot, are not included.

Estimate your monthly cost

Choose where it runs, pick a database and a size, then adjust anything you need.

WHERE IT RUNS

DATABASE

SIZE PER NODE

Database nodes

3

vCPU per node

€16 per vCPU

2 vCPU

Memory per node

€7 per GiB

8 GiB
100 GB

€0.15 per GB, usable capacity

200 GB

€0.05 per GB, across two regions

Estimated at twice your data

FULLY MANAGED · POSTGRESQL

€319per month, excl. VAT

Plus €199 one-time setup

2 vCPU per node: 6 × €16
€96
8 GiB RAM per node: 24 × €7
€168
100 GB NVMe per node: 300 × €0.15
€45
200 GB retained backups across two regions
€10
Total
€319

This provides 100 GB usable database capacity, replicated across three nodes.

Adding another 100 GB to every node costs €45/month.

Without an SLA we monitor around the clock and fix platform incidents. Support by email or Slack is included at no extra charge during business hours (09:00 to 17:00 Amsterdam time (CET/CEST)), best effort with no guarantees. An SLA (8x5, 16x5 or 24x7) gives you hard guarantees on response times; one SLA covers all your Cloud Exit services.

Book a call

How it compares

The Small size from the calculator, three PostgreSQL nodes with 2 vCPU and 8 GiB each, priced at four other providers.

PROVIDER AND SETUPPER MONTHEU COMPANYSELF-HOSTED OPTIONNOTES

Cloud Exit, fully managed

3 nodes · 2 vCPU · 8 GiB · 100 GB NVMe each · 200 GB backups in two regions

€319

Yes
Yes

Monitoring, failover, patching and restore checks included; support SLAs on contract

DigitalOcean

Managed PostgreSQL Advanced Edition · primary and 2 standby nodes · 2 vCPU · 8 GiB each

from €349

No
No

Additional storage €0.10 per GiB

AWS RDS

RDS for PostgreSQL · 3 × db.m7g.large (2 vCPU · 8 GiB) in Frankfurt · 300 GB gp3

from €423

No
No

Backup storage above the free allowance and data transfer billed separately

Google Cloud SQL

Enterprise edition in Frankfurt · HA pair plus 1 read replica · 2 vCPU · 8 GiB · 100 GB SSD each

from €378

No
No

Backups, data transfer and IP addresses billed separately

ayedo

Managed PostgreSQL Zero Downtime · managed Kubernetes cluster · 3 × 2 vCPU · 8 GB VMs · Germany

from €820

Yes
Yes

PostgreSQL €499.95, cluster €49.95, 3 VMs at €89.95 with 80 GB disk each. 24/7 support extra

Cloud Exit, fully managed

€319

3 nodes · 2 vCPU · 8 GiB · 100 GB NVMe each · 200 GB backups in two regions

EU COMPANYYesSELF-HOSTED OPTIONYes

Monitoring, failover, patching and restore checks included; support SLAs on contract

DigitalOcean

from €349

Managed PostgreSQL Advanced Edition · primary and 2 standby nodes · 2 vCPU · 8 GiB each

EU COMPANYNoSELF-HOSTED OPTIONNo

Additional storage €0.10 per GiB

AWS RDS

from €423

RDS for PostgreSQL · 3 × db.m7g.large (2 vCPU · 8 GiB) in Frankfurt · 300 GB gp3

EU COMPANYNoSELF-HOSTED OPTIONNo

Backup storage above the free allowance and data transfer billed separately

Google Cloud SQL

from €378

Enterprise edition in Frankfurt · HA pair plus 1 read replica · 2 vCPU · 8 GiB · 100 GB SSD each

EU COMPANYNoSELF-HOSTED OPTIONNo

Backups, data transfer and IP addresses billed separately

ayedo

from €820

Managed PostgreSQL Zero Downtime · managed Kubernetes cluster · 3 × 2 vCPU · 8 GB VMs · Germany

EU COMPANYYesSELF-HOSTED OPTIONYes

PostgreSQL €499.95, cluster €49.95, 3 VMs at €89.95 with 80 GB disk each. 24/7 support extra

Prices checked in October 2026, on demand and excluding VAT. US-dollar prices are converted to euros at €1 = $1.118 (26 October 2026). Sources: DigitalOcean PostgreSQL pricing documentation (Advanced Edition, €116 per 8 GiB / 2 vCPU node, standbys at the same rate); RDS db.m7g.large on-demand rate for Frankfurt (€131 per month), with gp3 storage at the US East rate of €0.10 per GB, so the Frankfurt total is higher; Google Cloud SQL Enterprise in Frankfurt, where compute is priced about 19% above Google's Iowa list rates of €0.037 per vCPU-hour and €0.0063 per GiB-hour (Bytebase regional price table), HA instances are billed at double the rate, and SSD storage is estimated at €0.18 per GB, Google's Frankfurt SSD disk rate; ayedo figures from its public price list, using the htz-vm-intel-2c-8gb VM. Each setup is the closest available match, and providers price storage, backups and data transfer differently.

Resources and expansion

RESOURCE OR EXPANSIONPRICECHARGED AGAINST
CPU€16 per vCPUDedicated cores per node, no oversubscription
Memory€7 per GiBProvisioned capacity per node
100% NVMe database storage€0.15 per GBProvisioned usable capacity per node, datacenter NVMe
Backups across two regions€0.05 per GBAverage retained backup size; both regional copies included

Resource rates apply to initial capacity and subsequent expansions. Backup usage includes retained full backups, incrementals and recovery logs.

FAQ

Questions we get

Do we need people who know Kubernetes?

No. Running the platform is our job, and your team keeps working on your applications. If you later decide to run it yourselves, any engineer who knows these standard open-source tools can take over.

Are we tied to Cloud Exit?

No. Everything runs on standard open-source software. In your own infrastructure, the accounts, data and setup are in your name, and your team can take over the same environment without moving anything. When we host it, the same setup can run in your infrastructure or with another provider.

Who do we call when something breaks?

Us, for everything below your application: servers, databases, storage and backups. When we respond depends on your SLA: 8x5 business hours, 16x5 extended hours or 24x7 for critical services. One SLA covers all your Cloud Exit services. Your own application stays with your team.

What is an offline environment?

Also called air-gapped: an environment with no connection to the internet at all. Software and updates are brought in through a controlled route, and every access is logged. It suits systems that handle the most sensitive data.

Can we start small?

Yes. A common starting point is one non-critical system on our infrastructure while the rest stays where it is. That is also how most migrations begin.