Geo-Replication Is Now Generally Available

Your backups exist. That's good. But where do they exist?

We take reliability seriously. Our infrastructure is built for uptime, with redundant systems, proactive monitoring, and careful maintenance. But there are events that no amount of operational excellence can prevent: a fire that destroys a building, an earthquake that takes out an entire region, a construction crew that severs a major fiber route, or a flood that forces evacuation. These are rare events — but when they happen, they're catastrophic.

Geo-replication is your protection against the uncontrollable.

Today, we're announcing the general availability of Geo-Replication — automatic, continuous replication of your backups across multiple geographic regions. No manual syncing. No complex scripts. Just select how many copies you want, and we handle the rest.

What Is Geo-Replication?

Geo-replication creates additional copies of your backup data in physically separate data centers. When you enable it, your backups are automatically synchronized to servers in different regions, keeping them safe from localized disasters.

Here's what happens under the hood:

1. You choose a replication factor (1-3 additional copies)
2. We automatically select destination servers in different geographic regions
3. Your data syncs every 15 minutes using ZFS-based incremental transfers
4. Each region maintains a complete copy of your datastore

The result: your data exists in multiple places, ready to be accessed even if an entire region becomes unavailable due to circumstances beyond anyone's control.

Why You Might Want This

Let's be honest — not everyone needs geo-replication. Our standard service is reliable, and for many users that's all they need. If you're running a homelab or can wait out the rare event of a major regional incident, single-region storage works well.

But some situations call for extra protection against the truly unpredictable.

Protection Against Acts of Nature

No data center — ours or anyone else's — is immune to natural disasters. Floods, fires, earthquakes, and severe weather events can take out entire regions regardless of how well the infrastructure is built. These events are rare, but when they happen, geographic separation is the only real protection.

With geo-replication, your data survives even if an entire region is affected by a natural disaster.

Business Continuity for Critical Operations

If your business absolutely cannot wait for disaster recovery — even for events outside anyone's control — geo-replication provides a safety net. When a major fiber cut or regional incident occurs, you can fail over to another region and continue restores without delay.

Compliance Requirements

Many regulatory frameworks require geographic separation of backup data. Whether it's GDPR, SOC 2, or industry-specific regulations, having your data in multiple regions helps satisfy "offsite backup" requirements with verifiable, auditable proof.

The True 3-2-1 Strategy

The classic 3-2-1 backup rule calls for at least one copy stored offsite. But "offsite" in the same city doesn't help when an earthquake or flood affects the whole area. Geo-replication gives you truly geographically separated copies — the kind that survive regional disasters.

How It Works

When you enable geo-replication, our system handles the complexity for you.

Automatic Server Selection

You don't need to pick which servers your data replicates to. Our algorithm automatically selects destination servers based on:

• Geographic diversity — each copy goes to a different region when possible
• Available capacity — servers with more headroom are preferred
• Server health — only healthy, online servers are selected

For example, if your primary datastore is in ham1, enabling 2x replication might place copies in fra1-1 and fra1-2. You get maximum geographic separation without any manual configuration.

Incremental Synchronization

Your first sync transfers all your backup data. After that, only changes are synchronized. This keeps bandwidth usage low and ensures your replica stays current without transferring terabytes of unchanged data every time.

The sync runs every 15 minutes, giving you a maximum of 15 minutes of potential data loss in a disaster scenario. For most backup use cases, this is more than sufficient — your actual backups likely run daily or weekly anyway.

Built on ZFS

Under the hood, we use ZFS send/receive for data transfer. This is the same battle-tested technology that powers enterprise storage systems worldwide. It's efficient, reliable, and handles incremental transfers gracefully.

Enabling Geo-Replication

Setting it up takes about 30 seconds:

That's it. Your data starts syncing immediately.

You can monitor the sync status directly in the dashboard. Each destination shows its current state, progress during active syncs, and the timestamp of the last successful sync.

Pricing

Geo-replication is priced per TB, per copy, per month:

4 EUR per TB per copy per month (excl. VAT)

This is based on your datastore's allocated size, not actual usage. So a 500 GB datastore with 2x replication would cost approximately 4 EUR/month (2 copies × 0.5 TB × 4 EUR).

Before enabling, you can preview the exact cost in the configuration modal. We show you the monthly estimate including VAT based on your account settings.

Real-World Scenarios

Scenario 1: The MSP With Client SLAs

You manage backups for 50 clients. Your SLA promises 4-hour RTO for critical restores. A major regional disaster — a fire, a flood, a severed backbone cable — would mean calling every client to explain why their RTO is now "when the region recovers."

With geo-replication, you fail over to another region and meet your SLAs. Even in catastrophic circumstances, you deliver.

Scenario 2: The Compliance Audit

Your auditor asks: "Where is your backup data stored? What's your disaster recovery plan if that region is affected by a major incident?"

Without geo-replication: "It's in one data center. In a major disaster, we wait for recovery."

With geo-replication: "Our data is replicated across three geographic regions. Even if an entire region is affected by a natural disaster or major infrastructure failure, we can restore from another location."

The second answer demonstrates mature disaster planning. The first one gets flagged.

Scenario 3: The Homelab That Got Serious

You started with a simple Proxmox homelab. Now you're running critical services — maybe a small business, maybe important personal projects. You trust your backup provider, but the thought of a once-in-a-decade regional disaster still nags at you.

2x replication for a 1 TB datastore costs about 8 EUR/month (net). That's the price of knowing your data survives anything — even events nobody can prevent.

What Geo-Replication Doesn't Replace

Let's be clear about what this feature is and isn't.

It's not a replacement for regular backups. Geo-replication copies your backup data to multiple regions. You still need to actually create backups in the first place.

It's not real-time failover. If a major disaster affects your primary region, your data is safe in other regions, but there's no automatic DNS failover or instant cutover. You'd need to manually point your Proxmox hosts to the replica location for restores.

It's not a backup of your backups. All copies contain the same data. If you accidentally delete a backup from your datastore, that deletion will sync to all replicas within 15 minutes.

Getting Started

Geo-replication is available now for all datastores. No feature flags, no waitlist, no enterprise tier required.

To enable it:

1. Log into your dashboard
2. Select your datastore
3. Open Replication Settings
4. Choose your replication factor and confirm

Your first sync will start immediately. Depending on your datastore size, initial synchronization may take a while — the dashboard shows you progress in real time.

Have questions? Reach out to support. We're happy to help you design a backup strategy that matches your resilience requirements.