Merge pull request GoogleCloudPlatform#3334 from GoogleCloudPlatform/…

…release-candidate

Release 1.43.0

cmd/root.go

@@ -53,7 +53,7 @@ HPC deployments on the Google Cloud Platform.`,
 				logging.Fatal("cmd.Help function failed: %s", err)
 			}
 		},
-		Version:     "v1.42.0",
+		Version:     "v1.43.0",
 		Annotations: annotation,
 	}
 )

community/examples/fsi-montecarlo-on-batch.yaml

@@ -15,6 +15,10 @@
 ---
 blueprint_name: fsi-montecarlo-on-batch
 
+validators:
+- validator: test_apis_enabled
+  skip: true # skipping this validator, since "service-enablement" will take care of it.
+
 vars:
   project_id:  ## Set GCP Project ID Here ##
   deployment_name: fsimontecarlo

community/examples/hpc-build-slurm-image.yaml

@@ -23,7 +23,7 @@ vars:
   image_build_machine_type: n2d-standard-16
   build_from_image_family: hpc-rocky-linux-8
   build_from_image_project: cloud-hpc-image-public
-  build_from_git_ref: 6.8.5
+  build_from_git_ref: 6.8.6
   built_image_family: my-custom-slurm
   built_instance_image:
     family: $(vars.built_image_family)

community/examples/xpk-gke-a3-megagpu.yaml

@@ -33,6 +33,7 @@ deployment_groups:
     source: modules/network/vpc
     settings:
       subnetwork_name: xpk-gke-a3-megagpu-subnet
+      mtu: 8244
       secondary_ranges:
         xpk-gke-a3-megagpu-subnet:
         - range_name: pods
@@ -47,6 +48,7 @@ deployment_groups:
       global_ip_address_range: 192.169.0.0/16
       network_count: 8
       subnetwork_cidr_suffix: 24
+      mtu: 8244
 
   - id: gke_cluster
     source: modules/scheduler/gke-cluster
@@ -70,25 +72,40 @@ deployment_groups:
       name: $(vars.deployment_name)-gp-np-1
       group_placement_max_distance: 2
 
-  - id: a3_megagpu_pool_0
+  - id: system_pool
+    source: modules/compute/gke-node-pool
+    use: [gke_cluster]
+    settings:
+      name: system-pool
+      machine_type: "e2-standard-16"
+      autoscaling_total_min_nodes: 2
+      autoscaling_total_max_nodes: 8
+      initial_node_count: 2
+      zones: [$(vars.zone)]
+      host_maintenance_interval: PERIODIC
+    outputs: [instructions]
+
+  - id: a3_megagpu_pool_np_0
     source: modules/compute/gke-node-pool
     use: [gke_cluster, gpunets, group_placement_0]
     settings:
-      name: a3-megagpu-pool-0
+      name: a3-megagpu-pool-np-0  # xpk naming scheme
       machine_type: a3-megagpu-8g
       autoscaling_total_min_nodes: 2
+      autoscaling_total_max_nodes: 150
       initial_node_count: 2
       zones: [$(vars.zone)]
       host_maintenance_interval: PERIODIC
     outputs: [instructions]
 
-  - id: a3_megagpu_pool_1
+  - id: a3_megagpu_pool_np_1
     source: modules/compute/gke-node-pool
     use: [gke_cluster, gpunets, group_placement_1]
     settings:
-      name: a3-megagpu-pool-1
+      name: a3-megagpu-pool-np-1  # xpk naming scheme
       machine_type: a3-megagpu-8g
       autoscaling_total_min_nodes: 2
+      autoscaling_total_max_nodes: 150
       initial_node_count: 2
       zones: [$(vars.zone)]
       host_maintenance_interval: PERIODIC
@@ -115,4 +132,4 @@ deployment_groups:
     settings:
       apply_manifests:
       - source: $(ghpc_stage("xpk-gke-a3-megagpu-files"))/config-map.yaml.tftpl
-        template_vars: {name: "xpk-gke-a3-megagpu-resources-configmap", num_nodes: "4"}
+        template_vars: {name: "$(vars.deployment_name)-resources-configmap", num_nodes: "4"}

community/modules/compute/gke-topology-scheduler/manifests/topology-scheduler-scripts.yaml

@@ -83,13 +83,11 @@ data:
       node_labels = node['node_labels']
 
       if (
-          'cloud.google.com/gke-placement-group' in node_labels
-          and 'topology.gke.io/cluster' in node_labels
+          'topology.gke.io/cluster' in node_labels
           and 'topology.gke.io/rack' in node_labels
           and 'topology.gke.io/host' in node_labels
       ):
         return (
-            node_labels['cloud.google.com/gke-placement-group'],
             node_labels['topology.gke.io/cluster'],
             node_labels['topology.gke.io/rack'],
             node_labels['topology.gke.io/host'],
@@ -144,13 +142,6 @@ data:
         node_name = node.metadata.name
         node_labels = node.metadata.labels
 
-        if 'cloud.google.com/gke-placement-group' not in node_labels:
-          print(
-              f'Skipping node {node_name} because it does not have topology'
-              ' metadata'
-          )
-          continue
-
         skip_node = False
         if node.spec.taints is not None:
           for t in node.spec.taints:

community/modules/compute/htcondor-execute-point/gpu_definition.tf

@@ -38,6 +38,7 @@ locals {
     "a2-ultragpu-8g" = { type = "nvidia-a100-80gb", count = 8 },
     "a3-highgpu-8g"  = { type = "nvidia-h100-80gb", count = 8 },
     "a3-megagpu-8g"  = { type = "nvidia-h100-mega-80gb", count = 8 },
+    "a3-ultragpu-8g" = { type = "nvidia-h200-141gb", count = 8 },
     "g2-standard-4"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-8"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-12" = { type = "nvidia-l4", count = 1 },

community/modules/compute/htcondor-execute-point/versions.tf

@@ -25,6 +25,6 @@ terraform {
   }
 
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:htcondor-execute-point/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:htcondor-execute-point/v1.43.0"
   }
 }

community/modules/compute/mig/versions.tf

@@ -22,6 +22,6 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:mig/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:mig/v1.43.0"
   }
 }

community/modules/compute/pbspro-execution/README.md

@@ -102,7 +102,7 @@ No resources.
 | <a name="input_machine_type"></a> [machine\_type](#input\_machine\_type) | Machine type to use for the instance creation | `string` | `"c2-standard-60"` | no |
 | <a name="input_metadata"></a> [metadata](#input\_metadata) | Metadata, provided as a map | `map(string)` | `{}` | no |
 | <a name="input_name_prefix"></a> [name\_prefix](#input\_name\_prefix) | Name prefix for PBS execution hostnames | `string` | `null` | no |
-| <a name="input_network_interfaces"></a> [network\_interfaces](#input\_network\_interfaces) | A list of network interfaces. The options match that of the terraform<br/>network\_interface block of google\_compute\_instance. For descriptions of the<br/>subfields or more information see the documentation:<br/>https://registry.terraform.io/providers/hashicorp/google/latest/docs/resources/compute_instance#nested_network_interface<br/><br/>**\_NOTE:\_** If `network_interfaces` are set, `network_self_link` and<br/>`subnetwork_self_link` will be ignored, even if they are provided through<br/>the `use` field. `bandwidth_tier` and `enable_public_ips` also do not apply<br/>to network interfaces defined in this variable.<br/><br/>Subfields:<br/>network            (string, required if subnetwork is not supplied)<br/>subnetwork         (string, required if network is not supplied)<br/>subnetwork\_project (string, optional)<br/>network\_ip         (string, optional)<br/>nic\_type           (string, optional, choose from ["GVNIC", "VIRTIO\_NET"])<br/>stack\_type         (string, optional, choose from ["IPV4\_ONLY", "IPV4\_IPV6"])<br/>queue\_count        (number, optional)<br/>access\_config      (object, optional)<br/>ipv6\_access\_config (object, optional)<br/>alias\_ip\_range     (list(object), optional) | <pre>list(object({<br/>    network            = string,<br/>    subnetwork         = string,<br/>    subnetwork_project = string,<br/>    network_ip         = string,<br/>    nic_type           = string,<br/>    stack_type         = string,<br/>    queue_count        = number,<br/>    access_config = list(object({<br/>      nat_ip                 = string,<br/>      public_ptr_domain_name = string,<br/>      network_tier           = string<br/>    })),<br/>    ipv6_access_config = list(object({<br/>      public_ptr_domain_name = string,<br/>      network_tier           = string<br/>    })),<br/>    alias_ip_range = list(object({<br/>      ip_cidr_range         = string,<br/>      subnetwork_range_name = string<br/>    }))<br/>  }))</pre> | `[]` | no |
+| <a name="input_network_interfaces"></a> [network\_interfaces](#input\_network\_interfaces) | A list of network interfaces. The options match that of the terraform<br/>network\_interface block of google\_compute\_instance. For descriptions of the<br/>subfields or more information see the documentation:<br/>https://registry.terraform.io/providers/hashicorp/google/latest/docs/resources/compute_instance#nested_network_interface<br/><br/>**\_NOTE:\_** If `network_interfaces` are set, `network_self_link` and<br/>`subnetwork_self_link` will be ignored, even if they are provided through<br/>the `use` field. `bandwidth_tier` and `enable_public_ips` also do not apply<br/>to network interfaces defined in this variable.<br/><br/>Subfields:<br/>network            (string, required if subnetwork is not supplied)<br/>subnetwork         (string, required if network is not supplied)<br/>subnetwork\_project (string, optional)<br/>network\_ip         (string, optional)<br/>nic\_type           (string, optional, choose from ["GVNIC", "VIRTIO\_NET", "RDMA", "IRDMA", "MRDMA"])<br/>stack\_type         (string, optional, choose from ["IPV4\_ONLY", "IPV4\_IPV6"])<br/>queue\_count        (number, optional)<br/>access\_config      (object, optional)<br/>ipv6\_access\_config (object, optional)<br/>alias\_ip\_range     (list(object), optional) | <pre>list(object({<br/>    network            = string,<br/>    subnetwork         = string,<br/>    subnetwork_project = string,<br/>    network_ip         = string,<br/>    nic_type           = string,<br/>    stack_type         = string,<br/>    queue_count        = number,<br/>    access_config = list(object({<br/>      nat_ip                 = string,<br/>      public_ptr_domain_name = string,<br/>      network_tier           = string<br/>    })),<br/>    ipv6_access_config = list(object({<br/>      public_ptr_domain_name = string,<br/>      network_tier           = string<br/>    })),<br/>    alias_ip_range = list(object({<br/>      ip_cidr_range         = string,<br/>      subnetwork_range_name = string<br/>    }))<br/>  }))</pre> | `[]` | no |
 | <a name="input_network_self_link"></a> [network\_self\_link](#input\_network\_self\_link) | The self link of the network to attach the VM. | `string` | `"default"` | no |
 | <a name="input_network_storage"></a> [network\_storage](#input\_network\_storage) | An array of network attached storage mounts to be configured. | <pre>list(object({<br/>    server_ip             = string,<br/>    remote_mount          = string,<br/>    local_mount           = string,<br/>    fs_type               = string,<br/>    mount_options         = string,<br/>    client_install_runner = map(string)<br/>    mount_runner          = map(string)<br/>  }))</pre> | `[]` | no |
 | <a name="input_on_host_maintenance"></a> [on\_host\_maintenance](#input\_on\_host\_maintenance) | Describes maintenance behavior for the instance. If left blank this will default to `MIGRATE` except for when `placement_policy`, spot provisioning, or GPUs require it to be `TERMINATE` | `string` | `null` | no |

community/modules/compute/pbspro-execution/variables.tf

@@ -203,7 +203,7 @@ variable "network_interfaces" {
     subnetwork         (string, required if network is not supplied)
     subnetwork_project (string, optional)
     network_ip         (string, optional)
-    nic_type           (string, optional, choose from ["GVNIC", "VIRTIO_NET"])
+    nic_type           (string, optional, choose from ["GVNIC", "VIRTIO_NET", "RDMA", "IRDMA", "MRDMA"])
     stack_type         (string, optional, choose from ["IPV4_ONLY", "IPV4_IPV6"])
     queue_count        (number, optional)
     access_config      (object, optional)

community/modules/compute/schedmd-slurm-gcp-v5-node-group/gpu_definition.tf

@@ -38,6 +38,7 @@ locals {
     "a2-ultragpu-8g" = { type = "nvidia-a100-80gb", count = 8 },
     "a3-highgpu-8g"  = { type = "nvidia-h100-80gb", count = 8 },
     "a3-megagpu-8g"  = { type = "nvidia-h100-mega-80gb", count = 8 },
+    "a3-ultragpu-8g" = { type = "nvidia-h200-141gb", count = 8 },
     "g2-standard-4"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-8"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-12" = { type = "nvidia-l4", count = 1 },

community/modules/compute/schedmd-slurm-gcp-v5-node-group/versions.tf

@@ -22,7 +22,7 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v5-node-group/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v5-node-group/v1.43.0"
   }
   required_version = ">= 1.1"
 }

community/modules/compute/schedmd-slurm-gcp-v5-partition/versions.tf

@@ -22,7 +22,7 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v5-partition/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v5-partition/v1.43.0"
   }
   required_version = ">= 0.13.0"
 }

community/modules/compute/schedmd-slurm-gcp-v6-nodeset-dynamic/README.md

@@ -74,7 +74,7 @@ modules. For support with the underlying modules, see the instructions in the
 
 | Name | Source | Version |
 |------|--------|---------|
-| <a name="module_slurm_nodeset_template"></a> [slurm\_nodeset\_template](#module\_slurm\_nodeset\_template) | github.com/GoogleCloudPlatform/slurm-gcp.git//terraform/slurm_cluster/modules/slurm_instance_template | 6.8.5 |
+| <a name="module_slurm_nodeset_template"></a> [slurm\_nodeset\_template](#module\_slurm\_nodeset\_template) | github.com/GoogleCloudPlatform/slurm-gcp.git//terraform/slurm_cluster/modules/slurm_instance_template | 6.8.6 |
 
 ## Resources
 

community/modules/compute/schedmd-slurm-gcp-v6-nodeset-dynamic/gpu_definition.tf

@@ -38,6 +38,7 @@ locals {
     "a2-ultragpu-8g" = { type = "nvidia-a100-80gb", count = 8 },
     "a3-highgpu-8g"  = { type = "nvidia-h100-80gb", count = 8 },
     "a3-megagpu-8g"  = { type = "nvidia-h100-mega-80gb", count = 8 },
+    "a3-ultragpu-8g" = { type = "nvidia-h200-141gb", count = 8 },
     "g2-standard-4"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-8"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-12" = { type = "nvidia-l4", count = 1 },

community/modules/compute/schedmd-slurm-gcp-v6-nodeset-dynamic/main.tf

@@ -56,7 +56,7 @@ locals {
 }
 
 module "slurm_nodeset_template" {
-  source = "github.com/GoogleCloudPlatform/slurm-gcp.git//terraform/slurm_cluster/modules/slurm_instance_template?ref=6.8.5"
+  source = "github.com/GoogleCloudPlatform/slurm-gcp.git//terraform/slurm_cluster/modules/slurm_instance_template?ref=6.8.6"
 
   project_id          = var.project_id
   region              = var.region

community/modules/compute/schedmd-slurm-gcp-v6-nodeset-dynamic/versions.tf

@@ -24,6 +24,6 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-nodeset-dynamic/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-nodeset-dynamic/v1.43.0"
   }
 }

community/modules/compute/schedmd-slurm-gcp-v6-nodeset-tpu/versions.tf

@@ -18,6 +18,6 @@ terraform {
   required_version = ">= 1.3"
 
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-nodeset-tpu/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-nodeset-tpu/v1.43.0"
   }
 }

community/modules/compute/schedmd-slurm-gcp-v6-nodeset/gpu_definition.tf

@@ -38,6 +38,7 @@ locals {
     "a2-ultragpu-8g" = { type = "nvidia-a100-80gb", count = 8 },
     "a3-highgpu-8g"  = { type = "nvidia-h100-80gb", count = 8 },
     "a3-megagpu-8g"  = { type = "nvidia-h100-mega-80gb", count = 8 },
+    "a3-ultragpu-8g" = { type = "nvidia-h200-141gb", count = 8 },
     "g2-standard-4"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-8"  = { type = "nvidia-l4", count = 1 },
     "g2-standard-12" = { type = "nvidia-l4", count = 1 },

community/modules/compute/schedmd-slurm-gcp-v6-nodeset/versions.tf

@@ -24,6 +24,6 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-nodeset/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-nodeset/v1.43.0"
   }
 }

community/modules/compute/schedmd-slurm-gcp-v6-partition/versions.tf

@@ -18,6 +18,6 @@ terraform {
   required_version = ">= 1.3"
 
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-partition/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:schedmd-slurm-gcp-v6-partition/v1.43.0"
   }
 }

community/modules/database/slurm-cloudsql-federation/versions.tf

@@ -26,10 +26,10 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:slurm-cloudsql-federation/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:slurm-cloudsql-federation/v1.43.0"
   }
   provider_meta "google-beta" {
-    module_name = "blueprints/terraform/hpc-toolkit:slurm-cloudsql-federation/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:slurm-cloudsql-federation/v1.43.0"
   }
 
   required_version = ">= 0.13.0"

community/modules/file-system/cloud-storage-bucket/versions.tf

@@ -26,7 +26,7 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:cloud-storage-bucket/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:cloud-storage-bucket/v1.43.0"
   }
   required_version = ">= 0.14.0"
 }

community/modules/file-system/nfs-server/versions.tf

@@ -30,7 +30,7 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:nfs-server/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:nfs-server/v1.43.0"
   }
 
   required_version = ">= 0.14.0"

community/modules/files/fsi-montecarlo-on-batch/versions.tf

@@ -35,9 +35,9 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:fsi-montecarlo-on-batch/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:fsi-montecarlo-on-batch/v1.43.0"
   }
   provider_meta "google-beta" {
-    module_name = "blueprints/terraform/hpc-toolkit:fsi-montecarlo-on-batch/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:fsi-montecarlo-on-batch/v1.43.0"
   }
 }

community/modules/network/private-service-access/README.md

@@ -88,6 +88,7 @@ No modules.
 
 | Name | Description |
 |------|-------------|
+| <a name="output_cidr_range"></a> [cidr\_range](#output\_cidr\_range) | CIDR range of the created google\_compute\_global\_address |
 | <a name="output_connect_mode"></a> [connect\_mode](#output\_connect\_mode) | Services that use Private Service Access typically specify connect\_mode<br/>"PRIVATE\_SERVICE\_ACCESS". This output value sets connect\_mode and additionally<br/>blocks terraform actions until the VPC connection has been created. |
 | <a name="output_private_vpc_connection_peering"></a> [private\_vpc\_connection\_peering](#output\_private\_vpc\_connection\_peering) | The name of the VPC Network peering connection that was created by the service provider. |
 | <a name="output_reserved_ip_range"></a> [reserved\_ip\_range](#output\_reserved\_ip\_range) | Named IP range to be used by services connected with Private Service Access. |

community/modules/network/private-service-access/outputs.tf

@@ -36,3 +36,8 @@ output "reserved_ip_range" {
   description = "Named IP range to be used by services connected with Private Service Access."
   value       = google_compute_global_address.private_ip_alloc.name
 }
+
+output "cidr_range" {
+  description = "CIDR range of the created google_compute_global_address"
+  value       = "${google_compute_global_address.private_ip_alloc.address}/${google_compute_global_address.private_ip_alloc.prefix_length}"
+}

community/modules/network/private-service-access/versions.tf

@@ -30,11 +30,11 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:private-service-access/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:private-service-access/v1.43.0"
   }
 
   provider_meta "google-beta" {
-    module_name = "blueprints/terraform/hpc-toolkit:private-service-access/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:private-service-access/v1.43.0"
   }
 
   required_version = ">= 1.2"

community/modules/project/service-enablement/versions.tf

@@ -22,7 +22,7 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:service-enablement/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:service-enablement/v1.43.0"
   }
 
   required_version = ">= 0.14.0"

community/modules/pubsub/bigquery-sub/versions.tf

@@ -26,10 +26,10 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:bigquery-sub/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:bigquery-sub/v1.43.0"
   }
   provider_meta "google-beta" {
-    module_name = "blueprints/terraform/hpc-toolkit:bigquery-sub/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:bigquery-sub/v1.43.0"
   }
   required_version = ">= 1.0"
 }

community/modules/pubsub/topic/versions.tf

@@ -27,6 +27,6 @@ terraform {
     }
   }
   provider_meta "google" {
-    module_name = "blueprints/terraform/hpc-toolkit:topic/v1.42.0"
+    module_name = "blueprints/terraform/hpc-toolkit:topic/v1.43.0"
   }
 }