PolarDB:Clone a cluster

Use cases

• Testing and development: Clone a production cluster to test schema migrations, parameter changes, or new application versions without impacting live traffic.

• Data analysis: Run intensive analytical queries or data exports on the clone instead of the production cluster.

• Disaster recovery drills: Validate your recovery procedures on a cloned cluster before an actual incident occurs.

• Staging environments: Create a staging environment that starts with a realistic copy of production data.

What gets cloned

The following table summarizes what is cloned for each edition. Only data present in the source cluster before cloning starts is included — data written after cloning begins is not captured.

Limitations

• SSL settings are never cloned, regardless of edition.

• Parameter settings and whitelist settings are not cloned for Enterprise Edition clusters. They are cloned for Standard Edition clusters.

• The cloned cluster is created in the same region as the source cluster.

• The cloned cluster uses the same compatibility (for example, PostgreSQL 14), product version, series, and CPU architecture as the source cluster.

Prerequisites

Before you begin, make sure that you have:

• An active PolarDB cluster to clone from

• A virtual private cloud (VPC) and vSwitch in the target zone

Clone a cluster

1. Log on to the PolarDB console.

2. In the left navigation pane, click Clusters.

3. In the upper-left corner, select the region where the source cluster is deployed.

4. Find the source cluster. In the Actions column, click More > Clone Data to New Cluster.

5. On the configuration page, set the parameters described in the following table.

   Parameter	Description
   Product type	Select Subscription (pay upfront for compute; storage billed hourly) or Pay-as-you-go (both compute and storage billed hourly).
   Operation type	Keep the default value Clone an independent cluster.
   Region	Automatically set to the same region as the source cluster. No change needed.
   Primary zone	Select the zone within the region. Zones are independent physical areas with no significant performance differences between them. Place the cluster in the same zone as your Elastic Compute Service (ECS) instances to minimize latency, though cross-zone placement is also supported.
   Network type	Only virtual private cloud (VPC) is supported. No change needed.
   VPC and vSwitch	Select the VPC and vSwitch for the cluster. To connect to ECS instances over the internal network, the cluster and ECS instances must be in the same VPC. If you do not have a VPC, create a VPC and a vSwitch first.
   Compatibility	Automatically set to match the source cluster (for example, PostgreSQL 14). No change needed.
   Product version	Automatically set to match the source cluster. No change needed.
   Series	Automatically set to match the source cluster (for example, Cluster Edition). No change needed.
   Sub-series	Select Dedicated or General-purpose. Dedicated clusters exclusively use their allocated CPU and memory, providing more stable performance. General-purpose clusters share idle compute resources across clusters on the same server, which reduces cost.
   CPU architecture	Automatically set to match the source cluster. No change needed.
   Node specifications	Select the compute node specifications for the clone. Different specifications provide different maximum storage capacities and performance levels. See Compute node specifications for Enterprise Edition or Compute node specifications for Standard Edition.
   Number of nodes	Default: 2 (one primary node and one secondary node). After the cluster is created, add more nodes as needed — up to one primary node and 15 secondary nodes. See .
   Database proxy type	Default: Dedicated Enterprise Edition.
   Enable hot standby storage cluster	When enabled, PolarDB deploys a primary cluster and a hot standby storage cluster in the same region, each with three data replicas (six replicas total). This improves the Service Level Agreement (SLA). When disabled, only the primary cluster with three replicas is deployed at half the storage unit price, with a lower SLA.
   Storage type	Enterprise Edition: Choose PSL5 (a storage type supported in earlier versions of PolarDB; provides good performance and high reliability and availability) or PSL4 (uses smart-SSD compression at the physical SSD level to lower cost per unit of data, while maintaining controllable performance impact). The storage type cannot be changed after the cluster is created. See to choose. Standard Edition: Choose from PL0, PL1, PL2, or PL3 ESSD (higher PL = higher IOPS and throughput), or ESSD AutoPL disk (decouples IOPS from capacity so you can adjust IOPS independently, reducing total cost of ownership (TCO)). See ESSDs for a full performance comparison. Important When an ESSD disk reaches capacity, it becomes locked and read-only. Monitor storage usage and expand capacity before the disk fills up.
   Storage billing method	Pay-as-you-go (by capacity): Storage scales automatically with your data; you pay only for what you use. Subscription (by space): Pre-purchase a fixed amount of storage. This parameter is only available when Product type is set to Subscription. Pay-as-you-go clusters always use Pay-as-you-go (by capacity) storage.
   Storage capacity	Specify the pre-purchased storage for Subscription (by space) billing. Ranges from 50 GB to 500 TB (Enterprise Edition) in 10 GB increments. Only available when both Product type and Storage billing method are set to Subscription.
   Cluster name	Enter a name for the cluster. The name must be 2–256 characters and cannot start with http:// or https://. Leave blank to auto-generate a name.
   Subscription duration	Specify the subscription term. Only available when the Billing method of the source cluster is Subscription.
   Quantity	Number of clusters to create. Valid values: 1–50. Default: 1.

6. Read and accept the service agreement, then complete the purchase:

   • Pay-as-you-go: Click Buy Now.

   • Subscription: Click Buy Now, then on the Payment page confirm the order details and payment method, and click Purchase.

Result

After the purchase, cluster provisioning takes 1–5 minutes. The cloned cluster then appears in the Clusters list.

What's next

• Add or remove nodes — scale the cloned cluster to match your workload

• — choose the right storage type for your cost and performance requirements