Efficiently manage your Clusters using the Tensor One CLI. This guide covers how to start Clusters (individually or in bulk), stop or schedule a stop, terminate Clusters, and list all active Clusters.

​
Start a Cluster

To start a single Cluster, run: 
tensoronecli start cluster $TENSORONE_CLUSTER_ID

​
Start Multiple Clusters (Bulk Start)

Start multiple Clusters simultaneously with: 
tensoronecli create clusters \
  --name bulk-task \
  --gpuType "NVIDIA GeForce RTX 3070,NVIDIA GeForce RTX 3080" \
  --imageName "tensorone/your-image-name" \
  --containerDiskSize 10 \
  --volumeSize 0 \
  --mem 1 \
  --args "bash -c 'your-bash-command'"
This command creates up to 10 Clusters using the specified GPUs and runs a given bash command on each one.

​
Stop a Cluster

To stop a single Cluster: 
tensoronecli stop cluster $TENSORONE_CLUSTER_ID

​
Stop a Cluster After a Set Duration

To automatically stop a Cluster after a specific amount of time (e.g., 2 hours): 
sleep 2h; tensoronecli stop cluster $TENSORONE_CLUSTER_ID &
  • sleep 2h waits for 2 hours
  • & runs the stop command in the background
Note: Idle Clusters still incur storage charges. Be sure to terminate Clusters when persistent storage is no longer needed.

​
Terminate a Cluster

Warning: Terminating a Cluster permanently deletes all data not stored on a network volume. Ensure you back up anything critical. Terminate a single Cluster with: 
tensoronecli remove cluster $TENSORONE_CLUSTER_ID
Terminate multiple Clusters (up to 40) by name: 
tensoronecli remove clusters my-bulk-task --clusterCount 40

​
List Active Clusters

To list all active Clusters: 
tensoronecli get cluster
This displays a summary of all currently active or available Clusters. You’re now ready to efficiently manage your Clusters using the full power of tensoronecli.