Scale a Cluster
Scale a cluster by changing its node size, number of nodes, or cache capacity. Choose a scaling mode based on your workload and capacity requirements.
Choose a scaling mode
| Your goal | Scaling mode | Behavior |
|---|---|---|
| Make a one-time change to node resources | Manual Scaling | Under Size, select Node Size and, for EBS, Cache per Node. |
| Make a one-time change to cluster size | Manual Scaling | Under Number, set Number of Nodes. |
| Adjust cluster size on a schedule | Scheduled Scaling | Set target node counts for selected days of the week and execution times. |
| Adjust cluster size as workload demand changes | Auto Scaling | Automatically adjust the number of nodes between configured minimum and maximum values. |
| Increase cache capacity without adding vCPU | Manual Scaling | For EBS, increase Cache per Node under Size. Local NVMe SSD has fixed cache capacity per node. |
Select a mode in Scaling Mode to configure it. For Auto Scaling, click Enable to start automatic adjustments or Disable to stop them. Disable Auto Scaling before using Manual Scaling or Scheduled Scaling.
Understand resource changes
Current Configuration shows Node Size, Cache per Node, and Number of Nodes. Total cluster vCPU, RAM, and cache capacity depend on the per-node configuration and the number of nodes.
Scheduled Scaling and Auto Scaling change the number of nodes. To change the resources of each node, use Manual Scaling.
Local NVMe SSD is available only for BYOC warehouses on AWS. Its cache capacity per node is fixed, so you cannot expand it independently like EBS cache.
Warning:
- Scaling can temporarily interrupt connections and delay requests.
- Reducing the number of nodes reduces total compute and cache capacity. Queries may need to read data from remote storage to refill the remaining cache.
- Local NVMe cache is cleared when a node is replaced.
- Increasing node size, node count, or EBS cache capacity can increase charges.
Before you scale
- You need permission to manage compute resources in the warehouse.
- The cluster must be Running, and the warehouse must not be in the Creating state.
- If Scaling is unavailable, point to it to view the reason.
- Review the target configuration and estimated cost before confirming a change.
Open scaling settings
- Log in to the VeloDB Cloud console.
- In the upper-left corner, select the warehouse that contains the cluster.
- In the left navigation pane, under Compute, click Clusters.
- Click Physical Clusters.
- Click Card, then select the running cluster to display its details panel.
- Click Scaling.
- If a resource menu opens, select On-Demand (Hourly).
In Scaling Mode, review Current Configuration, and then select Manual Scaling, Scheduled Scaling, or Auto Scaling.
After a resource change is submitted, the cluster status changes to Scaling and then returns to Running. Verify the node size, cache per node, and number of nodes in the cluster details panel.