# How to Migrate from Azure blob storage to Quake AI

Source: https://docs.quake.ai/docs/object/migration/migrate-from-azure
Markdown: https://docs.quake.ai/docs/object/migration/migrate-from-azure.md

---

# How to migrate from Azure blob storage to Quake AI

This guide walks through migrating objects from an Azure Blob Storage container to Quake AI object storage using rclone for direct cloud-to-cloud transfer.

## Service mapping

<MigrationTable provider="azure" service="storage/object" />


## Prerequisites

- A Quake AI account with [S3 credentials](/docs/object/how-to/create-s3-credentials)
- An Azure Storage Account SAS token with Read and List permissions, or an Azure service principal with **Storage Blob Data Reader** role
- [rclone](https://rclone.org/install/) installed (v1.65+)

## Before you start

Read the [migration planning guide](/docs/object/migration/plan-your-migration) to review S3/Swift feature compatibility and build a validation checklist.

**Azure-specific considerations:**

- **Terminology mapping**: Azure uses "containers" (equivalent to S3 buckets) and "blobs" (equivalent to S3 objects). Quake AI uses both "containers" and "buckets" interchangeably.
- **Access tiers**: Azure Blob offers Hot, Cool, Cold, and Archive tiers, plus a smart tier option that automatically moves data across the Hot, Cool, and Cold tiers based on access patterns (it does not cover Archive). Data on Quake AI is stored on a single tier. Two Azure tier behaviors affect migration timing and cost:
  - **Minimum retention and early deletion**: Azure sets a minimum retention period for cooler tiers: 30 days for Cool, 90 days for Cold, and 180 days for Archive. If you delete source objects after migrating them, objects still inside their minimum period incur an early deletion fee. Check each object's tier and age before you clean up the source.
  - **Archive rehydration**: Objects in the Archive tier must be rehydrated before they can be copied. Standard-priority rehydration can take up to 15 hours; high-priority rehydration can finish in under 1 hour for objects under 10 GB but costs more. Rehydration incurs a per-GB fee.
- **RBAC and ABAC**: Azure uses Entra ID (Azure AD) with RBAC and ABAC policies. These do not translate to Swift ACLs. Re-apply access controls on Quake AI using [grant access control](/docs/object/how-to/grant-access-control).
- **Lifecycle management**: Azure lifecycle policies (tier transitions, expiration) are not supported on Quake AI. See [retention without lifecycle policies](/docs/object/migration/plan-your-migration#retention-without-lifecycle-policies).
- **Hierarchical namespace (ADLS Gen2)**: If your Azure Storage Account has hierarchical namespace enabled (Azure Data Lake Storage Gen2), rclone's `azureblob` backend supports it but may require `adls = true` in the remote config, and SAS token scoping can differ. Verify your settings against the [rclone Azure Blob documentation](https://rclone.org/azureblob/).

## Configure rclone remotes

### Using a SAS token

Generate a SAS token in the Azure portal: **Storage Account** > **Shared access signature** > grant Read and List > generate.

Add both remotes to `~/.config/rclone/rclone.conf`:

```ini
[azure]
type = azureblob
account = YOUR_STORAGE_ACCOUNT_NAME
sas_url = https://YOUR_ACCOUNT.blob.core.windows.net/?YOUR_SAS_TOKEN

[quakeai]
type = s3
provider = Ceph
access_key_id = YOUR_RUMBLE_ACCESS_KEY
secret_access_key = YOUR_RUMBLE_SECRET_KEY
endpoint = object.us-east-2.rumble.cloud
acl = private
```

### Using a service principal

```ini
[azure]
type = azureblob
account = YOUR_STORAGE_ACCOUNT_NAME
tenant = YOUR_TENANT_ID
client_id = YOUR_CLIENT_ID
client_secret = YOUR_CLIENT_SECRET
```

Verify both remotes:

```bash
rclone lsd azure:
rclone lsd quakeai:
```

## Create the destination bucket

```bash
rclone mkdir quakeai:my-bucket
```

## Run the migration

### Initial copy

```bash
rclone copy azure:my-source-container quakeai:my-bucket \
  --transfers 16 \
  --checkers 32 \
  --s3-chunk-size 64M \
  --progress \
  --log-file migration.log \
  --log-level INFO
```

### Delta syncs

```bash
rclone sync azure:my-source-container quakeai:my-bucket \
  --transfers 16 \
  --checkers 32 \
  --progress
```

Schedule daily until cutover.

## Estimate egress cost

Azure charges per-GB for egress, with rates that vary by region. Azure does not currently run a free egress program for migrations. For current rates, see the [Azure bandwidth pricing page](https://azure.microsoft.com/en-us/pricing/details/bandwidth/).

For large data sets, consider a phased migration or check if your Azure support plan includes migration credits.

## Verify the migration

```bash
rclone size azure:my-source-container
rclone size quakeai:my-bucket
```

Check file integrity:

```bash
rclone check azure:my-source-container quakeai:my-bucket \
  --one-way \
  --log-file check.log
```

See the [validation checklist](/docs/object/migration/plan-your-migration#validation-checklist) for the full procedure.

## Cut over

1. Run a final `rclone sync`.
2. Update your application's storage endpoint. See [update application endpoint](/docs/object/migration/update-application-endpoint).
3. Keep the Azure source container for 30 days as a rollback target.

## See also

- [Plan your migration](/docs/object/migration/plan-your-migration): compatibility matrix, cost estimation, validation checklist
- [Migration tools comparison](/docs/object/migration/tools-comparison): rclone configuration and alternatives
- [Update application endpoint](/docs/object/migration/update-application-endpoint): swap SDK config to Quake AI