Using Infernet Cloud
Create an account
-
Navigate to Log in › Sign up.
-
Choose a name, email address, and password.
-
If email verification is enabled, verify your email address using the link sent to your email.
-
Log in.
Connect a cloud account
-
Navigate to Accounts › Connect Cloud Account.
-
Choose a cloud provider (AWS or Google Cloud), name your account, and click Continue.
-
Follow the instructions to install the cloud provider's CLI, run a script to generate credentials, and paste the credentials into the form.
Connect a Docker account to run private Docker images (optional)
-
Navigate to Accounts › Connect Docker account.
-
Follow the instructions to log into Docker Hub, generate credentials, and paste the credentials into the form.
Create container templates (optional)
-
Navigate to Templates › Create template.
-
Under "Container template name", choose a template name.
-
Under "Node configuration", specify the conditions needed for a node to run this container.
- To use a private Docker image, choose the owner's Docker Hub account.
-
Under "Container configuration", specify parameters needed to run the desired AI/ML workflow.
- To use a private Docker image, look for image IDs grouped under the owner's Docker Hub username.
-
Click Add container template.
Create a cluster
-
Navigate to Clusters › Create cluster.
-
Choose a connected cloud account to deploy the cluster with.
-
(Alternatively, navigate to Accounts › [Account] › Create cluster.)
-
-
Set a name, firewall rules, and default region and zone for the Infernet Router and Infernet Nodes.
-
Configure the location and machine for the Infernet Router (optional).
-
Configure one or more Infernet Nodes and their containers.
-
To add a container to a node, click "Add container". You can choose an existing container template or create a new one.
-
-
Click "Create cluster". It may take several minutes for the cloud resources to be provisioned.
Monitor and manage a cluster
-
Navigate to Clusters to view your clusters.
-
Click on a cluster to view its details, or right-click to trigger actions:
- Edit cluster: modify the configuration of the nodes (and router).
- Trigger update: manually trigger a redeployment of the cluster and provisioning of associated cloud resources.
- Destroy cluster: tear down the cluster and all associated nodes and cloud resources.
- Recreate cluster: recreate a destroyed cluster by reusing the existing configuration.
- Delete cluster: PERMANENTLY delete the configuration and deployment history of a destroyed cluster.
-
Scroll down to browse details about the cluster:
- Nodes: Infernet Nodes that are part of the cluster and their statuses.
- Router: configuration parameters for the Infernet Router (if the cluster was configured to deploy one).
- Configuration: configuration parameters for the cluster
- Deployment: logs, errors and created cloud resources from the last Terraform command run (representing the cluster's current state).
- History: all past Terraform commands triggered by Infernet Cloud and their inputs, outputs, and created cloud resources.
Monitor and manage nodes
-
From Clusters, navigate to a cluster detail page.
-
Click on a node to view its details, or right-click to trigger actions:
- Edit node: modify the node's configuration.
- Start node: start a stopped node.
- Stop node: stop a running node.
-
To add a node to an existing cluster, click "Add node".
-
Scroll down to browse details about the node:
- Containers: configured containers for this node.
- Status: status, IP and instance information for the node.
- Configuration: configuration parameters for the node.
- Logs: logs from the compute instance's serial port.