# Using standalone instances as AWS PCS login nodes
<a name="working-with_login-nodes_standalone"></a>

You can set up independent EC2 instances to interact with an AWS PCS cluster's Slurm scheduler. This is useful for creating login nodes, workstations, or dedicated workflow management hosts that work with AWS PCS clusters but operate outside of AWS PCS management. To do this, each standalone instance must:

1. Have a compatible Slurm software version installed.

1. Be able to connect to the AWS PCS cluster's Slurmctld endpoint.

1. Have the Slurm Auth and Cred Kiosk Daemon (`sackd`) properly configured with the AWS PCS cluster's endpoint and secret. For more information, see [sackd](https://slurm.schedmd.com/sackd.html) in the Slurm documentation.

 This tutorial helps you configure an independent instance that connects to an AWS PCS cluster. 

**Contents**
+ [Step 1 – Retrieve the address and secret for the target AWS PCS cluster](working-with_login-nodes_standalone_get-addr.md)
+ [Step 2 – Launch an EC2 instance](working-with_login-nodes_standalone_launch.md)
+ [Step 3 – Install Slurm on the instance](working-with_login-nodes_standalone_install-slurm.md)
+ [Step 4 – Retrieve and store the cluster secret](working-with_login-nodes_standalone_get-secret.md)
+ [Step 5 – Configure the connection to the AWS PCS cluster](working-with_login-nodes_standalone_configure-connection.md)
+ [Step 6 – (Optional) Test the connection](working-with_login-nodes_standalone_test.md)