The schedmd-slurm-gcp-v6-login module lets you create a login node for a Slurm
cluster. By using this module, you can provide a more secure entry point so that
users can do the following:
- Interact with the cluster
- Submit jobs
- Manage workloads
This module operates alongside the schedmd-slurm-gcp-v6-controller
module to establish a complete
Slurm environment.
For the complete list of inputs and outputs for this module, see the
schedmd-slurm-gcp-v6-login module
page in the Cluster Toolkit GitHub repository.
Before you begin
Before you begin, verify that you meet the following requirements:
- You have installed and configured Cluster Toolkit. For installation instructions, see Set up Cluster Toolkit.
- You have an existing cluster blueprint. You can use and modify an existing
blueprint or create one from scratch. For a working example of a blueprint
configured for Slurm, see the
examples/hpc-slurm.yamlfile. For more information about creating and customizing blueprints, see Cluster blueprint. - To view a complete list of blueprints that support Slurm, go to the Cluster blueprint catalog page, click the Select scheduler menu and then select Slurm.
- The
schedmd-slurm-gcp-v6-loginmodule does not create a full active cluster. Instead, it provisions a virtual machine (VM) instance that functions as a login node for interacting with the cluster and managing workloads.
Required roles
To get the permissions that you need to deploy the Slurm login node, ask your administrator to grant you the following IAM roles on your project:
- Compute Instance Admin (v1) (
roles/compute.instanceAdmin.v1) - Service Account User (
roles/iam.serviceAccountUser)
For more information about granting roles, see Manage access to projects, folders, and organizations.
You might also be able to get the required permissions through custom roles or other predefined roles.
Create a Slurm login node
The following example demonstrates how to configure a Slurm login node that connects to a specific subnetwork and associates with a Slurm controller. The example in this section uses the following configuration:
- Connects to the primary subnetwork of the
networkmodule. - Associates with the
slurm_controllermodule to establish the controller relationship. - Provisions a virtual machine (VM) instance that uses the
n2-standard-4machine type.
- id: slurm_login
source: community/modules/scheduler/schedmd-slurm-gcp-v6-login
use:
- network
settings:
group_name: login
machine_type: n2-standard-4
- id: slurm_controller
source: community/modules/scheduler/schedmd-slurm-gcp-v6-controller
use:
- network
- slurm_login
What's next
- For the complete list of inputs and outputs for this module, see the
schedmd-slurm-gcp-v6-loginmodule page in the Cluster Toolkit GitHub repository.