This topic describes the cluster specification change feature and its common scenarios.
PolarDB allows you to dynamically change the specifications of a cluster within just a few minutes. You do not need to lock databases when you change the specifications of the cluster. The cluster specification change feature supports the following three types of scaling:
Scale-up or scale-down of computing capacity: The specifications of the cluster are upgraded or downgraded. You can change the specifications of the primary and read-only nodes separately. This means that the specifications of read-only nodes do not have to be the same as those of the primary node.
NotePolarDB clusters provide a more flexible specification change solution. The specifications of read-only nodes can be changed separately and do not need to be the same as the specifications of the primary node. You can specify different specifications for different nodes based on your business requirements.
Scale-in or scale-out of computing capacity: Read-only nodes are added or removed. A maximum of 16 computing nodes can be added.
Scale-in or scale-out of storage capacity: The storage capacity that is billed by using the pay-as-you-go method can be scaled in serverless mode. The storage capacity is automatically adjusted as the storage usage increases or decreases. The storage capacity that is billed by using the subscription method can only be scaled manually. If you need to store a large volume of data, we recommend that you purchase storage plans to reduce costs. For more information, see Purchase a storage plan.
Cluster specification changes support the following scenarios:
Manually change specifications: You can manually change the specifications of your PolarDB for MySQL clusters. For more information, see Manually change the specifications of a cluster.
Automatically change specifications (auto-scaling): You can specify the parameters of the automatic specification change feature. PolarDB for MySQL Cluster Edition Edition instances automatically scale out and scale in based on the parameters that you configured. For more information, see Auto scaling for clusters that do not support serverless and Auto-scaling of local resources.
Temporary cluster upgrade: To improve the overall performance, you can temporarily upgrade the specifications of PolarDB clusters. When the rollback time is reached, the cluster is automatically rolled back to the original specifications. For more information, see Perform a temporary cluster upgrade.
NoteOnly subscription PolarDB clusters support temporary upgrades.
Add and delete nodes: You can manually add or delete read-only nodes. For more information, see Add or remove read-only nodes.
Upgrade Archive Database Single Node Edition to Archive Database Cluster Edition: You can upgrade existing Archive Database Edition clusters to X-Engine Edition Dedicated clusters. For more information, see Upgrade an Archive Database Standalone Edition cluster to an Archive Database Cluster Edition cluster.
Manually scale the storage capacity of a cluster: For more information, see Manually scale up or scale down storage capacity.