Managing Batch Jobs

This chapter covers how to manage jobs on DGX Cloud Lepton. For information on creating a new job, refer to Creating a DGX Cloud Lepton Job.

Job List

Visit the Jobs page to see a list of all jobs. You can filter jobs by status or by creator.

Explore Job Details

To view a job's details, click on its name in the job list. This will take you to the details page, where you can view the job status, duration, configurations, and more.

Replicas

Under the Replicas tab, you can see the status and DNS name of each replica. You can also view logs and metrics for each replica.

The replicas number represents the total number of workers in the job, which can be configured during job creation.

Metrics

The Metrics tab provides insights into the job's performance, including GPU usage, memory consumption, and GPU temperature.

Timeline

The Timeline tab offers a visual chart of the job's lifecycle, highlighting its progress across different stages.

Logs

The Logs page shows real-time logs for the job. You can also filter logs by time range for more targeted analysis.