Purchase a PolarDB for PostgreSQL cluster from the PolarDB console by selecting your edition, billing method, compute specifications, storage, and network configuration. Some settings cannot be changed after purchase—this guide highlights those decisions so you can configure the cluster correctly the first time.
Prerequisites
Before you begin, ensure that you have:
Procedure
Enterprise Edition
Step 1: Configure cluster parameters
Go to the custom purchase page for PolarDB clusters and configure the following parameters.
Billing method
| Option | Description | Suitable for | Changeable after purchase |
|---|---|---|---|
| Subscription | Pay upfront for a defined duration. Longer durations get larger discounts. | Long-term, stable workloads | Yes |
| Pay-as-you-go | Pay based on actual usage duration, with no upfront commitment. | Flexible or short-term workloads | Yes |
Region
Select the geographic location for your cluster.
You cannot change the region after purchase. Make sure the PolarDB cluster and the ECS instance you plan to connect to are in the same region. If they are in different regions, they cannot communicate over the internal network, which degrades performance.
For a list of available regions, see Regions and zones.
Creation method
| Option | Description |
|---|---|
| Create Primary Cluster | Creates a new PolarDB cluster. |
| Restore From Recycle Bin | Creates a cluster by restoring a deleted cluster's backup from the Recycle Bin. Select the Original Version, Deleted Cluster, and Historical Backup to restore. |
| Migrate From RDS | Clones data from an RDS instance and maintains incremental synchronization. The PolarDB cluster is read-only before the final migration switchover, and binary logging is enabled by default. Select the Source RDS Version (PostgreSQL 9 through 16) and Source RDS Instance (read-only instances are excluded). For details, see Solution overview. |
Engine and version
Under Database Engine, select a PostgreSQL version. PolarDB for PostgreSQL Enterprise Edition supports:
PostgreSQL 16
PostgreSQL 15
PostgreSQL 14
PostgreSQL 11
Set Edition to Enterprise Edition. Series is fixed to Cluster Edition (Recommended).
Sub-series
| Option | Resource model | Best for |
|---|---|---|
| Dedicated | Each cluster has exclusive use of its allocated CPU and compute resources—no sharing with other clusters on the same server. | Workloads requiring consistent, predictable performance |
| General-purpose | Clusters on the same server share idle CPU resources, maximizing utilization. | Cost-sensitive workloads with variable demand |
CPU Architecture is fixed to X86.
Network and zone
Primary Zone: Select a zone for the cluster. Zones are independent physical areas within a region with no substantial performance differences. You can place the cluster and your ECS instance in the same zone or in different zones.
Network Type: Fixed to Virtual Private Cloud (VPC). Make sure 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.
If you already have a VPC that meets your needs, select it. Otherwise, use the default VPC and vSwitch:
Default VPC: Unique within your selected region. Uses a 16-bit CIDR block (e.g., 192.168.0.0/16), providing up to 65,536 private IP addresses. Does not consume your VPC quota.
Default vSwitch: Unique within your selected zone. Uses a 20-bit CIDR block (e.g., 192.168.0.0/20), providing up to 4,096 private IP addresses. Does not consume the vSwitch quota.
To create a custom VPC, see Create and manage a VPC.
High-availability mode
| Option | Architecture | Data replicas | Storage cost |
|---|---|---|---|
| Dual-zone (Hot Standby Storage Cluster Enabled) | Primary cluster + hot standby storage cluster in the same region | 6 total (3 per cluster) | Standard |
| Single-zone (Hot Standby Storage Cluster Disabled) | Primary cluster only | 3 | Half of dual-zone |
If you select Dual-zone, also select a Secondary Zone. Automatic Allocation is supported.
For details about the hot standby storage cluster, see High-availability Mode.
Specifications and proxy
Node Specifications: Select the compute node specifications. For a full list, see Enterprise Edition compute node specifications.
Number of Nodes: The default is two nodes (one read/write node and one read-only node). You can select one node (read/write only) or two nodes.
Storage
Storage Type: Select PSL5 or PSL4.
| Storage type | Description | Changeable after purchase |
|---|---|---|
| PSL5 | Established storage type with good performance, high reliability, and high availability. | No |
| PSL4 | Uses Alibaba Cloud's self-developed smart-SSD technology to compress and decompress data at the physical SSD level, reducing the storage price per unit of data while keeping the performance impact manageable. | No |
You cannot switch the storage type for an existing cluster. To change storage types, purchase a new cluster with the desired type and migrate your data. For a comparison, see How to choose between PSL4 and PSL5.
Storage Billing Method (available only when Billing Method is Subscription):
| Option | Description |
|---|---|
| Billed By Capacity (Pay-as-you-go) | Serverless storage billing. Storage scales automatically as your data grows, and you pay only for actual usage. No capacity selection is required at purchase. |
| Billed By Space (Subscription) | Pay upfront for a fixed storage capacity (10 GB to 500 TB, in 10 GB increments). |
For a comparison of these billing methods, see Storage billing methods.
Advanced options (optional)
Cluster Name: Choose automatic generation or enter a custom name. Requirements for custom names:
Cannot start with
http://orhttps://2 to 256 characters in length
Must start with an uppercase letter, a lowercase letter, or a Chinese character
Can contain digits, periods (.), underscores (_), or hyphens (-)
Resource Group: Select a resource group to organize related resources under your Alibaba Cloud account. A resource can belong to only one resource group. For details, see Resource groups and authorization.
Enable TDE: Transparent Data Encryption (TDE) encrypts the cluster's data files. This process is transparent to your applications but may cause a 5%–10% performance loss.
TDE cannot be disabled after it is enabled.
TDE requires the KMS service. After enabling TDE, configure a service role and authorize PolarDB to access KMS.
Step 2: Review and purchase
On the right side of the page, verify that Billing Method, Region, Deployment Mode, and Resource Group match your requirements. Then configure:
| Parameter | Description |
|---|---|
| Quantity | Number of clusters to purchase. You can create up to 50 clusters at a time, which is useful for batch scenarios such as launching multiple game servers. Your Alibaba Cloud account can have a maximum of 50 clusters. |
| Subscription Duration | (Subscription billing only) Select the subscription duration and whether to Enable Auto-renewal. Enable auto-renewal to avoid service interruptions from a missed renewal. |
Review the total cost and cost breakdown (compute specifications, storage, and network). Click Buy Now to confirm you have read and agree to the Terms of Service and any product-specific terms. On the Payment page, confirm the order, select a payment method, and click Pay.
Result
After purchase, the cluster takes 10 to 15 minutes to create. Find it in the cluster list when ready.
Standard Edition
Step 1: Configure cluster parameters
Go to the custom purchase page for PolarDB clusters and configure the following parameters.
Billing method
| Option | Description | Suitable for | Changeable after purchase |
|---|---|---|---|
| Subscription | Pay upfront for a defined duration. Longer durations get larger discounts. | Long-term, stable workloads | Yes |
| Pay-as-you-go | Pay based on actual usage duration, with no upfront commitment. | Flexible or short-term workloads | Yes |
| Serverless | Post-paid with no predefined resource specifications. Cluster capacity scales dynamically based on workload changes. | Fluctuating, uncertain, or unpredictable workloads | Yes |
Region
Select the geographic location for your cluster.
You cannot change the region after purchase. Make sure the PolarDB cluster and the ECS instance you plan to connect to are in the same region. If they are in different regions, they cannot communicate over the internal network, which degrades performance.
For a list of available regions, see Regions and zones.
Creation method
| Option | Description |
|---|---|
| Create Primary Cluster | Creates a new PolarDB cluster. |
| Restore From Recycle Bin | Creates a cluster by restoring a deleted cluster's backup from the Recycle Bin. Select the Original Version, Deleted Cluster, and Historical Backup to restore. |
| Migrate from RDS | Clones data from an RDS instance and maintains incremental synchronization. The PolarDB cluster is read-only before the final migration switchover, and binary logging is enabled by default. Select the Source RDS Version (PostgreSQL 9 through 16) and Source RDS Instance (read-only instances are excluded). |
Engine and version
Under Database Engine, select a PostgreSQL version. PolarDB for PostgreSQL Standard Edition supports:
PostgreSQL 16
PostgreSQL 15
PostgreSQL 14
Set Edition to Standard Edition.
Sub-series
| Option | Resource model | Best for |
|---|---|---|
| Dedicated | Each cluster has exclusive use of its allocated CPU and compute resources—no sharing with other clusters on the same server. | Workloads requiring consistent, predictable performance |
| General-purpose | Clusters on the same server share idle CPU resources, maximizing utilization. | Cost-sensitive workloads with variable demand |
CPU architecture
| Option | Description |
|---|---|
| X86 | Uses Intel processors and high-performance networks for strong performance and stability. Suited to enterprise applications with high computing demands. |
| Yitian ARM | Uses Alibaba Cloud's self-developed Yitian 710 processor and 25 GE intelligent high-speed network interface cards. Delivers strong computing capabilities with better cost-effectiveness for cloud-native workloads. |
Network and zone
Primary Zone: Select a zone for the cluster. Zones are independent physical areas within a region with no substantial performance differences. You can place the cluster and your ECS instance in the same zone or in different zones.
Network Type: Fixed to VPC. Make sure 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.
If you already have a VPC that meets your needs, select it. Otherwise, use the default VPC and vSwitch:
Default VPC: Unique within your selected region. Uses a 16-bit CIDR block (e.g., 192.168.0.0/16), providing up to 65,536 private IP addresses. Does not consume your VPC quota.
Default vSwitch: Unique within your selected zone. Uses a 20-bit CIDR block (e.g., 192.168.0.0/20), providing up to 4,096 private IP addresses. Does not consume the vSwitch quota.
To create a custom VPC, see Create and manage a VPC.
High-availability mode
| Option | Architecture | Data replicas | Storage cost |
|---|---|---|---|
| Dual-zone (Hot Standby Storage Cluster Enabled) | Primary cluster + hot standby storage cluster in the same region | 6 total (3 per cluster) | Standard |
| Single-zone (Hot Standby Storage Cluster Disabled) | Primary cluster only | 3 | Half of dual-zone |
If you select Dual-zone, also select a Secondary Zone. Automatic Allocation is supported.
For details about the hot standby storage cluster, see High-availability Mode.
Specifications and proxy
Node Specifications: Select the compute node specifications. For a full list, see Enterprise Edition compute node specifications.
Number of Nodes: The default is two nodes (one read/write node and one read-only node). You can select one node (read/write only) or two nodes.
Database Proxy Type: Determined by your sub-series selection.
| Proxy type | Sub-series | Description |
|---|---|---|
| Standard Enterprise Edition | General-purpose | Shares physical CPU resources and provides second-level resource scalability based on workload. |
| Dedicated Enterprise Edition | Dedicated | Uses exclusive physical CPU resources for more stable performance. |
Proxy Specifications: Configure from 4 to 64 cores (4 vCPUs to 64 vCPUs), or select Auto-matching.
Storage
Storage Type: Select an Enhanced SSD (ESSD) performance level. ESSDs use next-generation distributed block storage with 25 GE networking and RDMA technology, delivering up to 1 million random read/write IOPS per disk.
| Storage type | Performance vs. previous level | Best for |
|---|---|---|
| PL0 ESSD | Baseline | Low-traffic or development workloads |
| PL1 ESSD | 5x IOPS, ~2x throughput vs. PL0 | Standard production workloads |
| PL2 ESSD | ~2x IOPS and throughput vs. PL1 | High-performance production workloads |
| PL3 ESSD | Up to 10x IOPS, 5x throughput vs. PL2 | Scenarios requiring extremely high concurrent I/O and stable latency |
| ESSD AutoPL Disk | IOPS decoupled from capacity; flexible, on-demand adjustment | Reducing total cost of ownership (TCO) with variable I/O needs |
For ESSD performance details, see ESSDs.
When a disk's storage space is full, the disk becomes read-only (locked).
Storage Billing Method: Follows the cluster's billing method.
Storage Capacity: Select the ESSD capacity to purchase (20 GB to 64 TB, in 10 GB increments).
Advanced options (optional)
Cluster Name: Choose automatic generation or enter a custom name. Requirements for custom names:
Cannot start with
http://orhttps://2 to 256 characters in length
Must start with an uppercase letter, a lowercase letter, or a Chinese character
Can contain digits, periods (.), underscores (_), or hyphens (-)
Resource Group: Select a resource group to organize related resources under your Alibaba Cloud account. A resource can belong to only one resource group. For details, see Resource groups and authorization.
Enable TDE: TDE encrypts the cluster's data files. This process is transparent to your applications but may cause a 5%–10% performance loss.
TDE cannot be disabled after it is enabled.
TDE requires the KMS service. After enabling TDE, configure a service role and authorize PolarDB to access KMS.
Step 2: Review and purchase
On the right side of the page, verify that Billing Method, Region, Deployment Mode, and Resource Group match your requirements. Then configure:
| Parameter | Description |
|---|---|
| Quantity | Number of clusters to purchase. You can create up to 50 clusters at a time, which is useful for batch scenarios such as launching multiple game servers. Your Alibaba Cloud account can have a maximum of 50 clusters. |
| Subscription Duration | (Subscription billing only) Select the subscription duration and whether to Enable Auto-renewal. Enable auto-renewal to avoid service interruptions from a missed renewal. |
Review the total cost and cost breakdown (compute specifications, storage, and network). Click Buy Now to confirm you have read and agree to the Terms of Service and any product-specific terms. On the Payment page, confirm the order, select a payment method, and click Pay.
Result
After purchase, the cluster takes 10 to 15 minutes to create. Find it in the cluster list when ready.
API reference
| API | Description |
|---|---|
| Create a cluster | Creates a database cluster. |
| Query a list of clusters | Retrieves a list of clusters. |
| View the detailed attributes of a PolarDB cluster | Retrieves the detailed attributes of a specified PolarDB cluster. |
| Query the auto-renewal status of a cluster | Queries the auto-renewal status of a subscription PolarDB cluster. |
| Configure the auto-renewal status for a subscription cluster | Configures the auto-renewal status for a subscription PolarDB cluster. |