--- title: "Change Unit settings" slug: "db-clouddbserverless-unit" updated: 2026-09-17T09:30:47Z published: 2026-09-17T09:30:47Z canonical: "guide.ncloud-docs.com/db-clouddbserverless-unit" stale: true --- > ## Documentation Index > Fetch the complete documentation index at: https://guide.ncloud-docs.com/llms.txt > Use this file to discover all available pages before exploring further. # Change Unit settings Available in VPC You can easily change the DB instance specification settings configured during the DB cluster creation process. Unit settings are applied uniformly to all instances in the selected DB cluster. Note - By default, DB management features are available when the DB instance status is **running**. - The available menus for DB management features differ depending on the DB cluster and DB instance. To change unit settings in the DB management menu: 1. In the VPC environment of the NAVER Cloud Platform console, navigate to **Menu** > **All Services** > **Database** > **Cloud DB Serverless**. 2. Click the **DB Cluster** menu. 3. In the DB cluster list, click the DB cluster you want to change, and then click **[DB Management]**. 4. In the **Change Unit Settings** menu, modify the minimum and maximum Unit information of the instance specifications. - Approximately 2 GB of memory is provided per Unit. - You can enter 1 to 165. 5. Change the setting and click **[Save]**. Note Cloud DB Serverless automatically generates adjustment rules for CPU utilization, connections, and available memory thresholds. Note The unit automatically increases or decreases based on CPU and memory usage. If there is high traffic and the CPU is used, the unit will increase according to usage up to the maximum unit you have set. Additionally, as the allocated unit increases, available memory expands, and parameters such as the buffer pool are automatically adjusted accordingly. Conversely, if traffic decreases, the unit will decrease according to usage to the minimum unit you have set, and related parameters will be adjusted accordingly.