Skip to content

Node monitoring

Now you should see your node in dashboard with status "Running". It means that you can run tasks on that agent. You can click on node name (generate automaticly) to see hardware info, "htop" and "nvidia-smi" (if available) commands output and other system information for node's health monitoring.

Open node dashboard

Go to "Cluster" page. Click to the name of interested node.

Node dashboard

  1. General information

  2. Local Agent storage information

  3. NVidia-smi

  4. htop

All information refreshes automatically every few seconds.

Manage local Agent storage

Removal operations are irreversible

Clean data carefully. Especially it is crutial for NN weights. Before remove NN weights folder check that all important models were uploaded to Supervisely Server.

User can clean cached images, stored NN weights and temporary task directory. Node storage shows actual information about used space.

To clean some of the directories just go to "Actions" and choose action you want.