> ## Documentation Index
> Fetch the complete documentation index at: https://bunnynet-cb9733c2-support-migration.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Durability

> Data durability guarantees, redundancy configuration, and disaster recovery capabilities

Bunny Storage achieves maximum data durability through multi-region redundancy and RAID 6 storage configuration.

## Durability guarantees

| Configuration                         | Durability                       | Annual Expected Loss    |
| ------------------------------------- | -------------------------------- | ----------------------- |
| Single region only                    | Standard                         | Variable                |
| With additional replication region(s) | **99.999999999%** (eleven nines) | 0.000000001% of objects |

<Info>
  Eleven nines durability (99.999999999%) corresponds to an average annual expected loss of 0.000000001% of objects.
</Info>

To achieve eleven nines durability, enable at least one additional replication region beyond your primary region. Each additional replication region further reduces the chance of data loss.

## Storage configuration

### RAID 6 configuration

Each datacenter stores a single replica of your data on **RAID 6 storage configuration**:

* Protects against simultaneous failure of up to two drives
* Provides redundancy at the hardware level
* Automatic rebuild when drives fail
* No data loss during drive failures

### Multi-region redundancy

When you enable replication regions:

* Data is automatically replicated across all selected regions
* Each region stores a complete copy of your data
* If data is lost in one datacenter, it is automatically re-streamed from another region
* Automatic failover ensures continuous availability

## Disaster recovery

Bunny Storage provides excellent durability even in extreme scenarios:

### Local disaster recovery

If a datacenter is damaged by natural disasters (earthquake, fire, flood, etc.):

* Automatic failover to the next closest region
* No data loss when using multiple replication regions
* Continuous availability for your users
* Transparent recovery with no action required

### Data loss protection

If data is lost in one datacenter:

* Data is automatically re-streamed from a different region
* New servers in the affected datacenter receive fresh copies
* No manual intervention required
* Replication continues in the background

### High availability

Your data remains accessible even during regional issues:

* Multiple regions provide redundancy
* Automatic routing to healthy regions
* No single point of failure
* 99.99% availability SLA

## Recommended configuration

For optimal durability and reliability:

### Minimum setup

* Enable at least **one additional replication region** for eleven nines durability
* Choose a region geographically distant from your primary region

### High availability setup

* Enable **multiple replication regions** across different continents
* Provides protection against regional disasters
* Ensures lowest latency for global users

### By storage tier

**Standard tier (HDD):**

* Enable geographically diverse regions (e.g., US, Europe, Asia)
* Up to 9 regions available

**Edge (SSD) tier:**

* Enable as many regions as possible for lowest latency
* Up to 15 regions available

See [Replication](/storage/replication) for details about configuring replication regions.

See [Pricing](/storage/pricing) for replication region costs.