Changing the CPU or Memory of a Cluster DB Instance

Scenarios

This section guides you on how to change the CPU or memory of a cluster instance.

  • A DB instance cannot be deleted when you are changing its CPU or memory.
  • Instances can be both scaled up and down.
  • Services will be interrupted for 5 to 10 minutes when you change DB instance CPU and memory. You are advised to perform this operation during off-peak hours.

Changing mongos

  1. Log in to the DDS console.
  2. On the Instance Management page, click the target cluster instance.
  3. In the Node Information area on the Basic Information page, click the mongos tab, locate the target mongos, and click Change Instance Class in the Operation column.
  4. On the displayed page, modify required parameters and click Submit.
  5. View the DB instance CPU or memory modification result.

    • When the CPU or memory of a DB instance is being changed, the status displayed in the Status column is Changing instance class. This process takes about 10 minutes.
    • In the upper right corner of the DB instance list, click to refresh the list. The instance status changes to Available.
    • In the Node Information area on the Basic Information page, click the mongos tab and check whether the scaling up is successful.

Changing shard

  1. Log in to the DDS console.
  2. On the Instance Management page, click the target cluster instance.
  3. In the Node Information area on the Basic Information page, click the shard tab, locate the target shard, and click Change Instance Class in the Operation column.
  4. On the displayed page, modify required parameters and click Submit.
  5. View the DB instance CPU or memory modification result.

    • When the CPU or memory of a DB instance is being changed, the status displayed in the Status column is Changing instance class. This process takes about 25 to 30 minutes.
    • In the upper right corner of the DB instance list, click to refresh the list. The instance status changes to Available.
    • Go to the Basic Information page of the cluster instance you scaled up, click the shard tab in the Node Information area, and check whether the scaling up is successful.

Parent topic: Changing DB Instance Classes