Skip to content

Instantly share code, notes, and snippets.

@nick-stroud
Created July 10, 2023 21:35
Show Gist options
  • Save nick-stroud/f866eacbcaaebf25479f9ccfe037ab78 to your computer and use it in GitHub Desktop.
Save nick-stroud/f866eacbcaaebf25479f9ccfe037ab78 to your computer and use it in GitHub Desktop.
Repro static node reconfigure
---
blueprint_name: dcluster
vars:
deployment_name: dcluster1
region: asia-southeast1
zone: asia-southeast1-c
project_id: # project id
enable_reconfigure: true
enable_cleanup_compute: true
deployment_groups:
- group: primary
modules:
- id: network1
source: modules/network/pre-existing-vpc
- id: homefs
source: modules/file-system/filestore
use: [network1]
settings:
local_mount: /home
- group: cluster
modules:
- id: tiny_node_group
source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
settings:
# create up to 4 cheap CPU nodes
node_count_dynamic_max: 4
node_count_static: 2
machine_type: n2-standard-2
- id: tiny_partition
source: community/modules/compute/schedmd-slurm-gcp-v5-partition
use:
- network1
- homefs
- tiny_node_group
settings:
partition_name: tiny
exclusive: false # allows nodes to stay up after jobs are done
enable_placement: false # the default is: true
# - id: c3_node_group
# source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
# settings:
# # create up to 4 cheap CPU nodes
# node_count_dynamic_max: 4
# node_count_static: 2
# machine_type: n2-standard-2
# - id: c3_partition
# source: community/modules/compute/schedmd-slurm-gcp-v5-partition
# use:
# - network1
# - homefs
# - c3_node_group
# settings:
# partition_name: c3
# exclusive: false # allows nodes to stay up after jobs are done
# enable_placement: false # the default is: true
- id: slurm_controller
source: community/modules/scheduler/schedmd-slurm-gcp-v5-controller
kind: terraform
use:
- network1
- homefs
- tiny_partition
# - c3_partition
settings:
instance_image:
family: slurm-gcp-5-7-debian-11
project: projects/schedmd-slurm-public/global/images/family
machine_type: "c2-standard-4" # this is the default value
disable_controller_public_ips: false
cloud_parameters:
no_comma_params: false
resume_rate: 0
resume_timeout: 300
suspend_rate: 0
suspend_timeout: 300
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment