forked from GoogleCloudPlatform/cluster-toolkit
-
Notifications
You must be signed in to change notification settings - Fork 0
/
hpc-cluster-small-sharedvpc.yaml
104 lines (93 loc) · 3.01 KB
/
hpc-cluster-small-sharedvpc.yaml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
# Copyright 2021 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.
---
blueprint_name: hpc-cluster-small-sharedvpc
# IMPORTANT NOTES
#
# 1. This blueprint expects a Shared VPC to exist and has already been shared
# from a Host project to a Service project.
# 2. It also anticipates that the custom steps for provisioning a Filestore
# instance in a Shared VPC in a service project have been completed:
#
# https://cloud.google.com/filestore/docs/shared-vpc
#
# 3. Replace project_id, host_project_id, network_name, subnetwork_name with
# valid values in your environment
vars:
project_id: ## Set GCP Project ID Here ##
host_project_id: your-host-project
network_name: your-shared-network
subnetwork_name: your-shared-subnetwork
deployment_name: hpc-small-shared-vpc
region: us-central1
zone: us-central1-c
deployment_groups:
- group: primary
modules:
- source: modules/network/pre-existing-vpc
kind: terraform
id: network1
settings:
project_id: $(vars.host_project_id)
- source: modules/file-system/filestore
kind: terraform
id: homefs
use: [network1]
settings:
local_mount: /home
project_id: $(vars.host_project_id)
connect_mode: PRIVATE_SERVICE_ACCESS
# This debug_partition will work out of the box without requesting additional GCP quota.
- source: community/modules/compute/SchedMD-slurm-on-gcp-partition
kind: terraform
id: debug_partition
use:
- network1
- homefs
settings:
partition_name: debug
max_node_count: 4
enable_placement: false
exclusive: false
machine_type: n2-standard-2
# This compute_partition is far more performant than debug_partition but may require requesting GCP quotas first.
- source: community/modules/compute/SchedMD-slurm-on-gcp-partition
kind: terraform
id: compute_partition
use:
- network1
- homefs
settings:
partition_name: compute
max_node_count: 20
- source: community/modules/scheduler/SchedMD-slurm-on-gcp-controller
kind: terraform
id: slurm_controller
use:
- network1
- homefs
- debug_partition # debug partition will be default as it is listed first
- compute_partition
settings:
login_node_count: 1
shared_vpc_host_project: $(vars.host_project_id)
- source: community/modules/scheduler/SchedMD-slurm-on-gcp-login-node
kind: terraform
id: slurm_login
use:
- network1
- homefs
- slurm_controller
settings:
shared_vpc_host_project: $(vars.host_project_id)