Preparations before table-level restore

Table-level restore requires the use of an auxiliary tenant. Therefore, before performing table-level restore, you need to create the necessary resource pools for the auxiliary tenant in the cluster where the target tenant resides.

Resource planning for the auxiliary tenant

During table-level restore, an auxiliary tenant must be created in the target cluster. Therefore, before starting table-level restore, you need to reserve sufficient CPU, memory, and disk resources in the cluster to prevent slow execution or failure of the table-level restore task due to insufficient resources.

Typically, it is recommended that the specifications for the auxiliary tenant match those of the source tenant. If this is not possible, you need to select appropriate specifications based on the system version.

In the current version, the system uses a fast restore method for table-level restore of the auxiliary tenant. The minimum specifications for the auxiliary tenant in fast restore mode are as follows:

Create a resource pool for the auxiliary tenant

1. Log in to the sys tenant of the cluster where the target tenant is located as the root user.

2. Create a unit (resource unit).

   Here is an example:

   obclient [(none)]> CREATE RESOURCE UNIT box_16c96g MAX_CPU 16, MEMORY_SIZE = '5G', MAX_IOPS 10240, MIN_IOPS=10240;
   

   For more information about how to create a resource unit, see Create a tenant.

3. Create a resource pool.

   When creating a resource pool for the auxiliary tenant, we recommend that you keep the architecture consistent with the source tenant. In other words, we recommend that the number of unit_num in the resource pool of the auxiliary tenant be the same as that of the source tenant.

   Here is an example:

   obclient [(none)]> CREATE RESOURCE POOL restore_pool unit = 'box_16c96g', unit_num = 1, zone_list = ('z1','z2','z3');
   

   For more information about how to create a resource pool, see Create a tenant.

What to do next

Perform table-level restore