You can view elastic jobs that are submitted by using Arena in AI Dashboard. This topic describes how to view the details of an elastic job in AI Dashboard.
Prerequisites
The Arena client is installed. For more information, see Install Arena.
The cloud-native AI suite is installed. For more information, see Deploy the cloud-native AI suite.
The credentials of the Resource Access Management (RAM) user that is specified as the administrator of AI Dashboard are obtained.
Run the following command to submit a training job by using Arena:
arena submit tf \ --name=tf-git \ --gpus=1 \ --image=kube-ai-registry.cn-shanghai.cr.aliyuncs.com/kube-ai/tensorflow:1.5.0-devel-gpu \ --sync-mode=git \ --sync-source=https://github.com/cheyang/tensorflow-sample-code \ "python code/tensorflow-sample-code/tfjob/docker/mnist/main.py --max_steps 10000 --data_dir=code/tensorflow-sample-code/data"Log on to AI Dashboard. For more information, see Access AI Dashboard.
In the left-side navigation pane of AI Dashboard, choose .
On the Training Job tab, you can view the training job submitted in Step 1. Click Detail in the Operator column to view the details of the training job.
On the Job Cost page, you can view the following information about the job:
The duration, estimated actual cost, estimated on-demand cost, and estimated saved cost.
The state, duration, resource type, instance type, and price of each pod.
You can submit inference jobs by using Arena or kubectl and then view the job details on the Inference Job tab.