Before proceeding, make a quick check that you have both clusters deployed. You should see two, one called “onprem” and one called “cloud”.
[ec2-user@server ~]$ pcluster list-clusters -r eu-west-1
{
"clusters": [
{
"clusterName": "onprem",
"cloudformationStackStatus": "CREATE_COMPLETE",
"cloudformationStackArn": "arn:aws:cloudformation:eu-west-1:196438911214:stack/onprem/c31c53e0-34d7-11ed-959e-02e9eca6697f",
"region": "eu-west-1",
"version": "3.5.1",
"clusterStatus": "CREATE_COMPLETE",
"scheduler": {
"type": "slurm"
}
},
{
"clusterName": "cloud",
"cloudformationStackStatus": "CREATE_COMPLETE",
"cloudformationStackArn": "arn:aws:cloudformation:eu-west-1:196438911214:stack/cloud/c31c53e0-34d7-11ed-959e-02e9eca6697f",
"region": "eu-west-1",
"version": "3.5.1",
"clusterStatus": "CREATE_COMPLETE",
"scheduler": {
"type": "slurm"
}
},
]
}
You may see other clusters that you have created earlier today, if you haven’t already cleaned up.
You should also see the database that has been pre installed for you.
[ec2-user@server ~]$ aws cloudformation --region ${AWS_REGION} describe-stacks --stack-name clusterdb | jq '.Stacks'[0]'.Outputs' | jq 'map(select(.OutputKey == "DatabaseHost"))'
[
{
"OutputKey": "DatabaseHost",
"OutputValue": "slurm-accounting-cluster.cluster-cshiwt8faqp5.eu-west-1.rds.amazonaws.com"
}
]
Assuming you can see the two HPC Clusters and the database it is ok to proceed. If not please ask one of the workshop staff.