As your business grows or shrinks—and as peak usage varies throughout the day—you need your database to deliver flexible service capabilities. OceanBase Database supports this elasticity with its tenant scaling feature.
Tenant scaling in and out lets you adjust computing power and storage capacity as needed. OceanBase Database provides two main ways to scale:
Increase the capacity of individual nodes: By raising the tenant's resource specifications (unit config), you can boost the performance of a single node.
For more information, see Tenant scaling in and out by adjusting resource specifications.
Add service nodes: By increasing the unit number and adding primary zones, you can expand the number of service nodes and enhance the tenant's overall performance. In simple terms, this scaling operation means expanding from N service units to M service units, where M is greater than N.
When a tenant's unit number changes from N to M, each primary zone needs to add M - N log streams, distributed across M - N newly added unit lists.
For more information, see Tenant scaling in and out by adjusting the unit number.
When the number of primary zones for a tenant changes from N to M, each unit list needs to add M - N log streams, with their leaders distributed across M - N newly added primary zones.
For more information, see Tenant scaling in and out by adjusting the number of primary zones.
Scaling in is simply the reverse of scaling out. Essentially, it reduces a tenant's service capabilities, including computing power and storage capacity. You can lower resource specifications to reduce the capacity of individual nodes, or decrease the number of service nodes—such as reducing the number of primary zones or unit numbers—to scale down the tenant's overall service performance.