Full restoration creates a new PolarDB cluster populated with all data from a historical backup set. The restore always targets a new cluster—it never overwrites the original. After you verify the restored data, you can migrate it back to the source cluster if needed.
The restored cluster inherits the data and account information from the source cluster, but not its parameter settings. Reconfigure any custom parameters on the new cluster after restoration.
Prerequisites
Before you begin, ensure that you have:
An existing PolarDB cluster with at least one backup set available
Sufficient account balance or subscription quota to create a new cluster
Restore from a backup set
Step 1: Open the Backup and Restoration page
Log on to the PolarDB console. In the left navigation pane, click Clusters.
Select the region where your source cluster resides, then click the cluster ID to open the cluster details page.
In the left navigation pane, choose Settings and Management > Backup and Restoration.
Step 2: Initiate the restore
Choose the restore scenario that applies:
Same-region restore: Find the target backup set in the list, then click Restore Data to New Cluster in the Actions column.
Cross-region restore:
On the Backup and Restoration page, select the region where the backup data resides.
Find the target backup set, then click Restore Data to New Cluster in the Actions column.
NoteCross-region restore is only available when cross-region backup is enabled on the source cluster. If cross-region backup is not enabled, the region is fixed to the source cluster's region.
Step 3: Configure the new cluster
On the Clone Instance page, select a Product Type for the new cluster:
Subscription: Pay for compute nodes upfront. Storage is billed hourly based on actual data volume.
Pay-as-you-go: No upfront payment. Both compute and storage are billed hourly. Use this option if you need the cluster temporarily—release it when done to avoid ongoing charges.
Then configure the following parameters:
| Parameter | Description |
|---|---|
| Operation type | Select Restore from Backup Set. |
| Region | Select the destination region. If cross-region backup is enabled, you can restore to either the source region or the destination region. |
| Backup set | Select the backup set to restore. Use the Backup Start Time shown for each backup set to identify the correct one. |
| Primary zone | Select the zone for the primary node. In regions with two or more zones, PolarDB automatically replicates data to a secondary zone for disaster recovery. |
| Network type | Fixed to VPC. No selection required. |
| VPC network | Select a virtual private cloud (VPC) and a vSwitch for the new cluster. We recommend that you select the same VPC and vSwitch as your source cluster. Ensure that the PolarDB cluster and the ECS instance you want to connect to are in the same VPC. Otherwise, they cannot communicate over the internal network for optimal performance. |
| VPC vSwitch | Configured together with VPC network. |
| Compatibility | Automatically set to match the original cluster's value (for example, PostgreSQL 14). No change required. |
| Edition | Automatically set to match the original cluster's product edition. No change required. |
| Series | Automatically set to match the original cluster's edition. No change required. |
| Sub-series | Select a specification type: Dedicated or General-purpose. Dedicated clusters exclusively use their allocated compute resources, such as CPU, without sharing them with other clusters on the same server, providing more stable and reliable performance. General-purpose clusters share idle compute resources across clusters on the same server, offering better cost-effectiveness. |
| CPU architecture | Automatically set to match the original cluster. No change required. |
| Node specifications | Select node specifications for the new cluster. Higher specifications support greater storage capacity and performance. To ensure the restored cluster performs as expected, select specifications higher than those of the original cluster. For details, see Product specifications. |
| Number of nodes | Defaults to 2 (one primary node and one secondary node). After the cluster is created, you can add up to 15 secondary nodes. See Add or remove nodes. |
| Database proxy type | Fixed to Dedicated Enterprise Edition. No selection required. |
| Enable Hot Standby Cluster | Enabling this option deploys a hot standby storage cluster alongside the primary cluster, for a total of six data replicas (three per cluster). This raises the Service-Level Agreement (SLA) ceiling. If you click Shut Down The Hot Standby Storage Cluster, only the primary cluster (with three data replicas) will be retained. The unit price for storage will be halved. The cluster's SLA will also be lowered. |
| Storage type | Depends on the cluster edition: Enterprise Edition supports PSL5 (higher performance and reliability) and PSL4 (uses Alibaba Cloud smart-SSD technology to reduce per-unit storage cost while keeping performance impact manageable). Standard Edition supports PL0, PL1, PL2, and PL3 ESSDs, and ESSD AutoPL disk. For a comparison, see How to choose between PSL4 and PSL5 and ESSD. You cannot change the storage type for an existing cluster. To change the storage type, purchase a new cluster with the desired storage type and migrate your data. |
| Storage billing type | Pay-as-you-go (billed by capacity): No upfront capacity selection needed. Storage scales automatically and you are charged only for actual usage. Subscription (billed by space): Prepay for a fixed storage capacity. When Product Type is Pay-as-you-go, storage defaults to pay-as-you-go and cannot be changed. See Storage billing methods. |
| Storage space | Applies only when Product Type is Subscription and Storage Billing Type is Subscription. For Enterprise Edition, the range is 50 GB–500 TB with a minimum increment of 10 GB. |
| Cluster name | Enter a name for the new cluster. Names must be 2–128 characters, start with a letter (uppercase or lowercase) or a Chinese character, and can include digits, periods (.), underscores (_), and hyphens (-). Leave blank to let the system generate a name automatically. You can rename the cluster after creation. |
| Subscription duration | Appears only when Product Type is Subscription. |
| Quantity | Number of clusters to create. |
Step 4: Complete the purchase
Pay-as-you-go: Click Buy Now to confirm you agree to the Terms of Service.
Subscription:
Click Buy Now to confirm you agree to the Terms of Service.
On the Payment page, confirm the order details and payment method, then click Pay.
After payment, cluster creation takes 10–15 minutes. When done, the new cluster appears on the Clusters page.
API reference
| API | Description |
|---|---|
| CreateDBCluster | Create a PolarDB cluster from a backup set. Set CreationOption to CloneFromPolarDB. |