Skip to content

Adding clarification to the networking in the "b. Create a distributed ML cluster" section #243

Description

@pptb-aws

In the section, "b. Create a distributed ML cluster", it has the user run these commands.

# create the cluster configuration
export IFACE=$(curl --silent http://169.254.169.254/latest/meta-data/network/interfaces/macs/)
export SUBNET_ID=$(curl --silent http://169.254.169.254/latest/meta-data/network/interfaces/macs/${IFACE}/subnet-id)
export VPC_ID=$(curl --silent http://169.254.169.254/latest/meta-data/network/interfaces/macs/${IFACE}/vpc-id)
export AZ=$(curl http://169.254.169.254/latest/meta-data/placement/availability-zone)
export REGION=${AZ::-1}

However, it does not provide an explanation regarding what these commands are for and what issues could arise based on these commands. I purposes adding the following information directly before this block.

"The following exports a number of variables that are used to set the VPC and subnet that the Head and Compute nodes will be launched into. These will use the VPC and Subnet of the instance where these commands are being run. Please review the networking requirments here for setting up Parallel Cluster in a single public subnet to make sure your subnet meets the requirments"

Additionally, there is a typo in the ml-config.yaml where, in the HeadNode.Dcv Enabled is spelled Emabled.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions