View Cloud Services and Manage Resources
You can view cloud services and manage compute resources in C3 AI Studio to inspect resource utilization, adjust resources to scale applications, and allocate resources appropriately.
In C3 AI Studio, a user with the StudioAdmin role can do the following on a cluster level:
- View cloud services and their details.
- View compute resources across environments and applications.
- Add node pools to manage workloads.
- Customize hardware profiles to optimize resource allocation and performance.
By default, users with the StudioUser role can view services for their own environments and apps.
Requirements
Have the StudioAdmin role.
View cloud services
Based on the services your C3 AI Studio instance is using, you may see the following cloud services:
- Cassandra
- Ex Machina
- PostgreSQL
- Spark
- Jupyter Notebook
Here is how to view cloud services in C3 AI Studio:
- In C3 AI Studio, navigate to the Admin page.
- Select Services. A list of all services and their details in the C3 AI Studio instance display.
View and manage compute resources
You can view compute resources, add node pools, and customize hardware profiles across your cluster in C3 AI Studio.
View compute resources
Here is how to view compute resources:
- In C3 AI Studio, navigate to the Admin page.
- Select Services.
- Select Compute. A list of node pools and their compute resources display.
Add a node pool
To learn more about node pools, see Configure and Manage Node Pools.
Here is how to add a node pool:
- In Services > Compute, select the + icon in the Node Pools panel.
- Provide the basic information, the number of nodes, and hardware profile information for the node pool.
- (Optional) Select Advanced to further configure storage, memory, and compute.
Customize a hardware profile
Here is how to customize a hardware profile:
- In Services > Compute, select the node pool that you want to configure a hardware profile for.
- Select Configure.
- Under Hardware Profile, set the CPU, Memory, GPU, and GPU Type for the node pool.