Provisioning an Oracle Exadata VM Cluster requires the existence of an Oracle Exadata Infrastructure, and is a prerequisite for Oracle Exadata Databases that runs on the cluster.

1. From Google Cloud console, open the Oracle Database@Google Cloud application.
2. From the left menu, select the Exadata Database menu item.
3. By default, the Oracle Exadata Infrastructure tab is selected. To create a Oracle Exadata VM Cluster resource, select that tab first.
4. Select the + CREATE icon.
5. In the Basic details section, enter the following information.
   1. Select the Oracle Exadata Infrastructure from the drop-down list. You must have completed the Provisioning Exadata Infrastructure steps in order to continue.
   2. Enter a Display name. It is required, and must be unique to your account.
   3. Enter a VM Cluster ID. It is required, and must be unique to your account. It can be the same value as the Display name.
   4. The Cluster name field is optional, and provides a name to ease management in large systems.
   5. Select the Oracle Grid Infrastructure version from the drop-down list of versions. The version selected here determines the Oracle Database versions the Oracle Exadata VM Clusters support.
6. In the VM cluster configuration section, enter the following information.
   1. DB servers for VM placement is a read-only field that shows you the servers available from the Oracle Exadata Infrastructure. There is a minimum of two (2) Database Servers. If your Oracle Exadata Infrastructure has more than two (2), you can select the servers you want to be used by selecting the CHANGE DB SERVERS button, selecting two (2) or more servers from the available list, and then selecting the SUBMIT button.
   2. Enter the number for OCPU count per VM, in the range from 0 to 126. If you change the value of OCPU count per VM, the read-only Total requested OCPU count field recalculates.
   3. Enter the number for Memory per VM, in the range from 30 to 1,390 GBs. If you change the value of Memory per VM, the read-only Total requested memory field recalculates.
   4. Enter the number for Local storage per VM, in the range from 60 to 1,800 GBs. If you change the value of Local storage per VM, the read-only Total local storage field recalculates.
7. In the Exadata storage configuration section, enter the following information.
   1. Enter the number for Usable Exadata Storage, in the range from 2 to 100 TBs in 1 TB increments.
   2. The Allocate storage for sparse snapshots checkbox can only be selected prior to provisioning. This options selects a sparse disc group, which is required to support Exadata snapshot functionality.
   3. The Allocate storage for local backups checkbox can only be selected prior to provisioning. This options enables local Exadata storage as a backup destination for any Oracle Database backups.
   4. The Usable storage allocation field is read-only and shows the distribution of Data, Recovery, and Sparse storage based on your selections.
8. In the Networking section, enter the following information.
   1. Select the Associated network from the drop-down list of available VPC networks.
   2. Enter the Client IP Range value in CIDR notation. Ensure that the CIDR range does not overlap with any other subnets in the selected VPC.
   3. Enter the Backup IP Range value in CIDR notation. Ensure that the CIDR range does not overlap with any other subnets in the selected VPC.
   4. Enter a Hostname prefix that will form the first portion of the Oracle Exadata VM Cluster host name.
9. In the SSH keys section, enter a public SSH key.
   1. You must enter at least one (1) public SSH key.
   2. You can enter multiple public SSH keys.
   3. SSH keys must use RSA encryption. You can use a command like, ssh-keygen -t rsa -f ~/.ssh/KEY_FILENAME -C USERNAME -b 2048 to generate a usable SSH key.
10. In the License type section, select either License included or Bring your own license (BYOL).
11. The Diagnostics collection section provides checkboxes to enable collection of diagnostics and notifications to you. You must subscribe to those events to receive the notifications. The available selections are as follows.
    1. The Enable diagnostic events checkbox allows Oracle to collect and send fault notifications about critical, warning, and information events.
    2. The Enable health monitoring checkbox allows Oracle to collect health metrics and events, such as Oracle Database up/down, disk space usage, and share them with Oracle Cloud operations.
    3. The Enable incident logs and trace collections checkbox allows Oracle to collect incident logs and traces to enable fault diagnosis and issue resolution.
12. By default, the Advanced fields section is collapsed. If you want to see the current values or change these values, open the Advanced fields section and enter the following information.
    1. Select the Timezone from the drop-down list of available timezones.
    2. By default, the SCAN Listener Port (TCP/IP) is 1521. You can enter a SCAN Listener Port in the range between 1024 and 8000.
13. You are now ready to provision your Oracle Exadata VM Cluster. Select the CREATE button to begin the provisioning process, or the CANCEL button to stop the process without provisioning.
14. Wait for the Oracle Exadata VM Cluster provisioning process to complete before attempting to provision any Oracle Exadata VM Clusters on the provisioned Oracle Exadata VM Clusters.
15. Open the details for the Oracle Exadata VM Cluster, and select the MANAGE IN OCI icon. The OCI console opens with your Oracle Exadata VM Cluster details shown.
16. Select the Create database button. For additional details, see Create Oracle Database Homes on an Exadata Cloud Infrastructure System.