Services Instances
The Services Instance page in the GPU PaaS portal provides a unified interface for creating and managing various GPU-accelerated service instances, including Inference, Notebook, Job, and Custom Services. By leveraging service profiles and compute instances, users can deploy GPU resources to match specific requirements, whether it's running AI/ML models, executing batch jobs, or hosting custom workloads. This flexibility ensures that diverse application needs are met efficiently, maximizing the utility of available GPU resources.
This page streamlines the deployment and management process, offering detailed insights into the status, type, and configuration of services. Users can monitor their services in real-time, make adjustments as needed, and ensure optimal performance for their workloads. The Service Instance page is an essential feature for users working on GPU-intensive tasks like deep learning, data processing, or custom computational workflows, enabling faster development and improved resource utilization.
Managed Service Instances¶
PaaS Admins can configure and launch service instances by following the steps described below.
- Access PaaS Studio and click on the Service Instances menu on the left to access the service instances page
- Click on New Service Instance. A prompt will appear to specify the following in a wizard-like experience.
Available Service Profiles¶
The list of service profiles available for use with service is displayed here. Choose the desired profile and fill in the necessary details
Unique Name¶
Provide a unique name for the service instance (e.g. deepcompute-notebook). Note that you cannot change the unique name once the profile has been created.
Display Name¶
This is typically a user friendly name such as "notebook service" that will make sense to the end users. This name is what the end users will see an experience.
Description¶
Provide a multiple sentence description that will help the end users understand more details about the service.
Service Profiles¶
When ready to proceed with the selected service profile, proceed with compute instance selection. Alternatively, click Change Service Profile to select a different profile.
Click Save & Continue
The Overview page appears with the inputs provided
Labels¶
Labels are optional metadata in the form of key-value pairs that will be automatically attached to the resulting service when they are launched. Labels help in categorizing and organizing services based on specific criteria like environment, project, or team.
Annotations¶
Annotations are also optional metadata in key-value pair format, but unlike labels, they are typically used for descriptive purposes rather than for categorization or filtering. Annotations provide additional context or information about the service.
- Users can either click Save Changes to save the configuration for later publishing, keeping the deployment status as Not Deployed
- Click Deploy to immediately proceed with publishing the changes
- On clicking Deploy, the service instance deployment process begins, and the status updates to In Progress
- To add or edit the labels and annotations, click the Edit Configuration button. This will direct to the Overview page, where changes to labels and annotations can be made
Note: Service Instance with the status In Progress cannot be destroyed. This restriction ensures that the deployment process is not interrupted or corrupted
Once publishing is successful, the deployment status changes to Success
View Services¶
When PaaS Admins click on the Service Instances menu on the left, a complete list of available service instances is displayed. They can search for a specific service instance by entering input into the search box. The main listing provides basic information about each service instance.
- Unique Name
- Service Type
- Creation Date
- Publish Status
- Sharing
- Actions Menu (ellipses)
To view additional details about a service instance, simply click on it. This will display all the details the administrator had configured during initial creation.
Update Services¶
Follow the steps above to "View" service instances. Alternatively, click on the actions (ellipses) on the far right of a service instance and select "Edit Configuration". PaaS Admins can now make necessary changes to the instances and save it.
Delete Services¶
To delete a service instance, click on the actions (ellipses) on the far right of a instance and select delete.
Info
This is a destruction action and cannot be reversed. Once an instance is deleted, it is no longer available to end users for use.
Share Services¶
To share a service instance with selected or all projects, click on the actions (ellipses) on the far right of a service profile and select "Manage Sharing". Now, select how you wish to share the project. The options are:
- None (This is the default)
- All Projects
- Selected Projects
The ability to share service instances with other projects allows platform teams to centralize the mgmt of service profiles and share them hierarchically with downstream projects where they can be used by end users in multiple teams/departments.
