You can create a new queue and specify the resources and operators required for running your AI task.

Prerequisites

• You have created a Kubernetes cluster successfully. For details, see Create Cluster.
• You already successfully installed the CCE AI Job Scheduler and CCE Deep Learning Frameworks Operator components. Otherwise, the cloud native AI feature is unavailable.

Restriction Description

• Before you allocate GPU resources to the queue, you need to install the CCE GPU Manager component.

Operation Steps

1. Log in to Baidu AI Cloud Official Website, and then enter the management console.
2. Select “Product Service > Cloud Native > CCE” and click CCE to enter the container engine console.
3. Click Cluster Management > Cluster List in the navbar on the left side.
4. On the cluster list page, click the target cluster name to enter the cluster management page.
5. On the cluster management page, click Cloud Native AI > Queue Management.
6. On the queue management page, click Create Queue.
7. On the new queue page, complete the basic information configuration of the queue:

• Queue name: customizes a queue name with a length of 1–63 characters, which can only contain numbers, lowercase letters "-" and must start and end with a letter or number.
• Remarks: The remarks of the custom queue should not exceed 300 characters in length.

8. Complete the configuration of the queue quota information:

• CPU: Specify the CPU resource size for the queue. The input limit is a number greater than 0, and the number is rounded to a maximum of two decimal places.
• Memory: Specify the memory resource size for the queue. The input limit is a number greater than 0, and the number is rounded to a maximum of two decimal places. Also, you can change the size unit between Gi and Mi.
• Exclusive GPU: Specify the GPU resource allocated by the whole card for the queue. The input limit is an integer greater than 0.
• Shared GPU: Specify the GPU resource allocated by the graphic memory for the queue. The input limit is an integer greater than 0.

Note: Before you specify the GPU resource for the queue, you need to install the CCE GPU Manager component.

9. Complete the resource object configuration.

• Object type: Specify the type of available resource object for the queue. At present, you can specify “user” only.
• Object selection: Specify the sub-users who can use the queue resources. You can select multiple sub-users.

10. Click the "OK" button to complete the queue creation.