AI Applications

This guide explains the procedure for managing AI applications.

AI Applications Table

The AI applications table can be found under Workload manager in the NVIDIA Run:ai platform.

The AI applications table provides a list of all the AI applications scheduled on the NVIDIA Run:ai Scheduler, and allows you to manage them.

The AI applications table consists of the following columns:

Column	Description
AI application	The name of the AI application
Type	The name of the Helm chart
Status	The different phases in an AI application lifecycle
Project	The project in which the AI application runs
GPU compute request	Amount of GPU devices requested
GPU compute allocation	Amount of GPU devices allocated
GPU memory request	Amount of GPU memory Requested
GPU memory allocation	Amount of GPU memory allocated
CPU compute request	Amount of CPU cores requested
CPU compute allocation	Amount of CPU cores allocated
CPU memory request	Amount of CPU memory requested
CPU memory allocation	Amount of CPU memory allocated

AI Application Status

The AI application status in NVIDIA Run:ai reflects the underlying Helm release status.

NVIDIA Run:ai surfaces the Helm chart state as-is and maps it to the AI application lifecycle. For a complete description of Helm release states, see the Helm documentation.

Customizing the Table View

• Filter - Click ADD FILTER, select the column to filter by, and enter the filter values

• Search - Click SEARCH and type the value to search by

• Sort - Click each column header to sort by

• Column selection - Click COLUMNS and select the columns to display in the table

• Download table - Click MORE and then Click Download as CSV. Export to CSV is limited to 20,000 rows.

• Refresh - Click REFRESH to update the table with the latest data

• Show/Hide details - Click to view additional information on the selected row

Show/Hide Details

Click a row in the AI applications table and then click the SHOW DETAILS button at the upper-right side of the action bar. The details pane appears, presenting detailed breakdown of some Kubernetes resources that belong to it. The details pane displays:

• A list of all AI application components (workloads, services, secrets, PVCs, ConfigMaps, etc.)

• Status indicators (Running, Pending, Failed, etc.) for each workload

Managing and Monitoring

After the AI application is created, the workloads are added to the Workloads table, where they can be managed and monitored.

Using API

Go to the AI Applications API reference to view the available actions.