Quick Start Guide
A comprehensive guide to getting started with AI Computing Platform, covering storage creation, file uploads, container instances, and environment adaptation.
Create Storage
- Log in to the management console.
- In the top navigation bar, click Products and Services > AI Computing Platform > AI Computing Platform to go to its overview page.
- In the left navigation bar, select Storage & Data Services to access the corresponding management page.
- On the Storage and Data Services page, click Create User Directory. In the Create User Directory window that pops up, enter the user directory, set the storage quota, and click OK.
Notice:
- The user directory name here must be globally unique and cannot be repeated.
- Each user can only create one user directory.
- Return to the Storage and Data Services Management page, and you can view the created file directory with the status of
normal.
Upload Files
- In the left navigation bar, select Storage & Data Services. In the File Directory area, select the directory where the local file is to be uploaded.
- Click Upload Local Files, drag the file to the upload file window, or click Upload.
- Close the upload file window, and the file has been uploaded to the specified folder.
Create Container Instance
- In the left navigation bar, select Container Instances. The container instance list page is displayed by default.
- Click Create Container Instance. On the Create Container Instance page that pops up, configure various parameters and click OK.
Parameter Configuration:
| Parameter | Explanation |
|---|---|
| Instance Name | User-defined name for easy identification. |
| Resource | - Billing mode: Default is pay-as-you-go. |
| - Resource type: Select high-speed training, shared GPU, or CPU computing types. Depending on the resource type, the GPU model, CPU model, memory, etc., will differ. | |
| Storage and Data (optional) | Select the user directory where the dataset is located and the corresponding mount directory. |
| Mirror | Select a public image, custom image, or private image address. |
| Public images: Pre-built images within the platform supporting TensorFlow, Pytorch, Jupyter, etc. | |
| Custom image: Image built by the user based on a preset image or Dockerfile. | |
| Image address: Public or user private image address. If a password is set, check the box with password and enter the username and password to obtain the corresponding image. |
- Return to the Container Instances page and wait for the container instance to be created. The successfully created instance is displayed with a status of
running.
Log in to the Container Instance and Adapt to the Environment
- Click Jupyter in the Quick Access column of the specified container instance to jump to the JupyterLab page. For detailed usage of JupyterLab, refer to JupyterLab Usage Introduction.
Notice:
- If you use a custom image to create a container instance, ensure that the image file contains the JupyterLab service. Otherwise, the JupyterLab page cannot be opened.
- If you cannot open the JupyterLab page, please check whether the pop-up window blocking function of your browser is disabled.
- In the JupyterLab workspace, click Other > Terminal to open the terminal and log in to the container instance backend.
- Users can execute the following command to view the system default source.