Help Center/ TaurusDB/ User Guide/ Read Replicas/ Adding Read Replicas to a DB Instance
Updated on 2025-08-21 GMT+08:00

Adding Read Replicas to a DB Instance

Scenarios

In read-intensive scenarios, a single DB instance may be unable to handle the read pressure and workloads may be affected. To offload read pressure from the primary node, you can create one or more read replicas. These read replicas can process a large number of read requests. Read replicas enhance read capabilities of your DB instance and reduce load on the primary node. After a DB instance is created, you can add read replicas.

Deployment Relationships Between the Primary Node and Read Replicas

  • If you select single-AZ deployment, read replicas are deployed in the same AZ as the primary node.
  • If you select multi-AZ deployment, read replicas are evenly deployed in different AZs to ensure high reliability.

Constraints

  • Each yearly/monthly or pay-per-use DB instance has a maximum of 15 read replicas.
  • Each serverless DB instance has a maximum of 7 read replicas.
  • If all synchronous read replicas are unavailable during a failover, an asynchronous read replica will be promoted to primary.

Billing

Table 1 Billing for new read replicas

Billing Mode of New Read Replicas

Impact on Price

Yearly/Monthly

You will be billed for the new read replicas based on the time remaining in the requested period of your instance.

You need to pay the price difference.

The following prices are for reference only. The actual prices are displayed on the console.

Suppose you purchased a one-month TaurusDB cluster instance (instance specifications: dedicated, 2 vCPUs | 8 GB, 2 nodes; storage: DL6, 10 GB) in CN-Hong Kong on April 1, 2025. The instance price was $296 USD per month.

On April 15, 2025, you added a read replica (specifications: dedicated, 2 vCPUs | 8 GB). The instance price was $436 USD per month.

Price difference = Price for the new instance configuration x Remaining period – Price for the original instance configuration x Remaining period

The remaining period is the remaining days of each calendar month divided by the maximum number of days in each calendar month.

In this example, the remaining period and price difference are calculated as follows: Remaining period = 15 (Remaining days in April)/30 (Maximum number of days in April) = 0.5. Price difference = $436 USD x 0.5 – $296 USD x 0.5 = $70 USD

Pay-per-use

New read replicas are billed by hour. For details, see Product Pricing Details.

Procedure

  1. Log in to the management console.
  2. Click in the upper left corner and select a region and project.
  3. Click in the upper left corner of the page and choose Databases > TaurusDB.
  4. On the Instances page, locate the instance you want to add read replicas to and choose More > Create Read Replica in the Operation column.

    You can also enter the Create Read Replica page in either of the following ways:

    • Click the instance name to go to the Basic Information page. In the upper right corner of the page, click and choose Create Read Replica.
      Figure 1 Creating read replicas on the Basic Information page
    • Click the instance name to go to the Basic Information page. In the upper right corner of the page, click and choose View Instance Topology. In the instance topology, click Create Read Replica.
      Figure 2 Creating read replicas in the instance topology

  5. On the displayed page, set required parameters.

    Table 2 Parameter description

    Parameter

    Description

    Billing Mode

    • Pay-per-use DB instance: Pay-per-use and serverless read replicas can be added.
    • Yearly/Monthly DB instance: Yearly/Monthly, pay-per-use, and serverless read replicas can be added.
    • Serverless DB instance: Only serverless read replicas can be added.

    Failover Priority

    Failover priority ranges from 1 for the first priority to 16 for the last priority. This priority determines the order in which read replicas are promoted when recovering from a primary node failure. Read replicas with the same priority have a same probability of being promoted to the new primary node. You can configure a failover priority for up to 9 read replicas, and the default priority for the remaining read replicas is -1, indicating these read replicas cannot be promoted to primary. You can change the failover priority of a read replica.

    • Serverless DB instance: The failover priority of the primary node can only be 1.
    • Yearly/Monthly DB instance: When a pay-per-use or serverless read replica is added, the failover priority is -1 by default and cannot be changed.
    • Pay-per-use DB instance: When a serverless read replica is added, the failover priority is -1 by default and cannot be changed.

    AZ

    TaurusDB multi-AZ instances allow you to select an AZ when creating a read replica. You cannot specify AZs for pay-per-use or yearly/monthly instances.

    • If no AZs are specified, the created read replicas are evenly distributed in each AZ.
    • If too many nodes are created in a specified AZ, the read replicas may fail to be created due to insufficient resources.
    NOTE:

    This function is in the OBT phase. To use it, submit a service ticket.

    Instance Specifications

    This parameter is only available for cluster instances.

    If the failover priority is set to 1, the specifications of read replicas must be the same as those of the primary node.

    Quantity

    A DB instance can contain up to 15 read replicas.

  6. For a yearly/monthly instance, click Next and select a payment mode.
  7. For a pay-per-use instance, click Next.
  8. Check the read replica settings.

    • If you need to modify the settings, click Previous.
    • If you do not need to modify the settings, click Submit.

  9. View the new read replicas in the Node List area of the Basic Information page. You can also promote a read replica to primary or delete a read replica.