slurm.yaml

# Copyright 2022 Google LLC
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#      http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Name    Spec                  Hourly Prices (standard/spot)
# debug   n1-standard-8 + P4    $0.35 / $0.32
# p100    n1-standard-8 + P100  $1.06 / $0.67
# g2      g2-standard-16 + L4   $1.15 / $0.5
# v100    n1-standard-8 + V100  $1.77 / $1.07
# a2      a2-highgpu-1g + A100  $3.67 / $1.47
---
blueprint_name: hpc-slurm-ubuntu2004
vars:
  project_id:  bentest-403619
  deployment_name: slurm-gcp-v5
  region: us-central1
  zone: us-central1-a
  # zones:
  # - us-central1-a
  # - us-central1-b
  # - us-central1-c
  instance_image:
    # Please refer to the following link for the latest images:
    # https://github.com/SchedMD/slurm-gcp/blob/master/docs/images.md#supported-operating-systems
    family: slurm-gcp-5-9-ubuntu-2004-lts
    project: schedmd-slurm-public
deployment_groups:
- group: primary
  modules:
  # Source is an embedded resource, denoted by "resources/*" without ./, ../, /
  # as a prefix. To refer to a local resource, prefix with ./, ../ or /
  # Example - ./resources/network/vpc
  - id: network1
    source: modules/network/vpc
  - id: homefs
    source: modules/file-system/filestore
    use: [network1]
    settings:
      size_gb: 1024
      local_mount: /home
  - id: debug_node_group
    source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
    settings:
      node_count_dynamic_max: 4
      machine_type: n1-standard-8
      guest_accelerator:
      - type: nvidia-tesla-t4
        count: 1
  - id: debug_partition
    source: community/modules/compute/schedmd-slurm-gcp-v5-partition
    use:
    - network1
    - homefs
    - debug_node_group
    settings:
      partition_name: debug
      exclusive: false # allows nodes to stay up after jobs are done
      enable_placement: false # the default is: true
      is_default: true
  - id: v100_node_group
    source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
    settings:
      node_count_dynamic_max: 20
      bandwidth_tier: gvnic_enabled
      machine_type: n1-standard-8
      guest_accelerator:
        - type: nvidia-tesla-v100
          count: 1
  - id: v100_partition
    source: community/modules/compute/schedmd-slurm-gcp-v5-partition
    use:
    - network1
    - homefs
    - v100_node_group
    settings:
      partition_name: v100
  - id: v100_preemptible_node_group
    source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
    settings:
      node_count_dynamic_max: 20
      bandwidth_tier: gvnic_enabled
      machine_type: n1-standard-8
      guest_accelerator:
        - type: nvidia-tesla-v100
          count: 1
      preemptible: true
  - id: v100_preemptible_partition
    source: community/modules/compute/schedmd-slurm-gcp-v5-partition
    use:
    - network1
    - homefs
    - v100_preemptible_node_group
    settings:
      partition_name: v100p
      enable_placement: false # the default is: true
  - id: g2_node_group
    source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
    settings:
      node_count_dynamic_max: 20
      bandwidth_tier: gvnic_enabled
      machine_type: g2-standard-16
  - id: g2_partition
    source: community/modules/compute/schedmd-slurm-gcp-v5-partition
    use:
    - network1
    - homefs
    - g2_node_group
    settings:
      partition_name: g2
  - id: a2_node_group
    source: community/modules/compute/schedmd-slurm-gcp-v5-node-group
    settings:
      node_count_dynamic_max: 20
      bandwidth_tier: gvnic_enabled
      machine_type: a2-highgpu-1g
  - id: a2_partition
    source: community/modules/compute/schedmd-slurm-gcp-v5-partition
    use:
    - network1
    - homefs
    - a2_node_group
    settings:
      partition_name: a2
      enable_placement: false # the default is: true
  - id: slurm_controller
    source: community/modules/scheduler/schedmd-slurm-gcp-v5-controller
    use:
    - network1
    - debug_partition
    - v100_partition
    - v100_preemptible_partition
    - g2_partition
    - a2_partition
    - homefs
    settings:
      disable_controller_public_ips: false
      cloud_parameters:
        resume_rate: 0
        resume_timeout: 600  # Update this value, default is 300
        suspend_rate: 0
        suspend_timeout: 300
        no_comma_params: false
  - id: slurm_login
    source: community/modules/scheduler/schedmd-slurm-gcp-v5-login
    use:
    - network1
    - slurm_controller
    settings:
      machine_type: n2-standard-2
      disable_login_public_ips: false