All Products
Search
Document Center

:Best practices for scaling

Last Updated:Mar 11, 2024

ApsaraDB for OceanBase allows you to modify instance configurations in a standard or temporary manner for instance scaling. This topic describes scaling methods in different scenarios.

Modify the storage space

You can modify the storage space to implement instance scaling. For more information, see Modify the storage space.

We recommend that you scale out your storage space in the following scenarios:

  • Your business keeps running and generates more data. More storage space is required.

  • The storage space is running out, or you have received the relevant alerts.

However, when you migrate or delete historical business data, more storage space is released. In this case, you can scale in the storage space.

Modify plan specifications

You can modify plan specifications to implement instance scaling. For more information, see Modify plan specifications.

When you need to increase the business performance, you can scale out your instance by adjusting plan specifications. When the business traffic returns to normal after peak hours, you can scale in your instance by adjusting plan specifications.

Add and delete a node

You can add or delete nodes to implement instance scaling. For more information, see Add and delete a node.

As your business keeps running and generates more data, the capacity of a cluster or tenant instance becomes insufficient. You can add more nodes. After you add one node to a cluster instance, one node is added to each zone of the cluster instance. If you need to reduce the computing capacity of a cluster instance, you can delete nodes.

Scale out the log disk

You can scale out the log disk to implement instance scaling. For more information, see Scale out the log disk.

If you need to prolong the log retention period and keep more logs, you can scale out the log disk. If you run a large-specification cluster instance that has many small-specification tenants, we recommend that you scale out the log disk.

Enable automatic storage scale-out

If your business involves a large data volume or your business volume frequently fluctuates drastically, we recommend that you enable automatic storage scale-out. When the storage space of a cluster instance is about to be exhausted, the storage space is automatically scaled out to ensure the stability of online services. For more information, see Enable automatic storage scale-out.

Modify tenant specifications

When the performance of a tenant instance does not meet the business requirements or the data traffic of the new business is low in a test environment, you can scale out or in the tenant instance by modifying the tenant specifications. For more information, see Modify tenant specifications.