made efa optional

ai-ml/jark-stack/terraform/README.md

@@ -20,6 +20,7 @@ Docs coming soon...
 | Name | Version |
 |------|---------|
 | <a name="provider_aws"></a> [aws](#provider\_aws) | >= 3.72 |
+| <a name="provider_aws.ecr"></a> [aws.ecr](#provider\_aws.ecr) | >= 3.72 |
 | <a name="provider_kubernetes"></a> [kubernetes](#provider\_kubernetes) | >= 2.10 |
 
 ## Modules
@@ -41,13 +42,16 @@ Docs coming soon...
 | [kubernetes_namespace_v1.jupyterhub](https://registry.terraform.io/providers/hashicorp/kubernetes/latest/docs/resources/namespace_v1) | resource |
 | [kubernetes_secret_v1.huggingface_token](https://registry.terraform.io/providers/hashicorp/kubernetes/latest/docs/resources/secret_v1) | resource |
 | [kubernetes_storage_class.default_gp3](https://registry.terraform.io/providers/hashicorp/kubernetes/latest/docs/resources/storage_class) | resource |
+| [aws_availability_zones.available](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/data-sources/availability_zones) | data source |
+| [aws_ecrpublic_authorization_token.token](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/data-sources/ecrpublic_authorization_token) | data source |
 | [aws_eks_cluster_auth.this](https://registry.terraform.io/providers/hashicorp/aws/latest/docs/data-sources/eks_cluster_auth) | data source |
 
 ## Inputs
 
 | Name | Description | Type | Default | Required |
 |------|-------------|------|---------|:--------:|
 | <a name="input_eks_cluster_version"></a> [eks\_cluster\_version](#input\_eks\_cluster\_version) | EKS Cluster version | `string` | `"1.29"` | no |
+| <a name="input_enable_aws_efa_k8s_device_plugin"></a> [enable\_aws\_efa\_k8s\_device\_plugin](#input\_enable\_aws\_efa\_k8s\_device\_plugin) | Enable AWS EFA K8s Device Plugin | `bool` | `false` | no |
 | <a name="input_huggingface_token"></a> [huggingface\_token](#input\_huggingface\_token) | Hugging Face Secret Token | `string` | `"DUMMY_TOKEN_REPLACE_ME"` | no |
 | <a name="input_name"></a> [name](#input\_name) | Name of the VPC and EKS Cluster | `string` | `"jark-stack"` | no |
 | <a name="input_region"></a> [region](#input\_region) | region | `string` | `"us-west-2"` | no |

ai-ml/jark-stack/terraform/addons.tf

@@ -202,11 +202,16 @@ module "data_addons" {
   #---------------------------------------
   # EFA Device Plugin Add-on
   #---------------------------------------
-  enable_aws_efa_k8s_device_plugin = true
+  # IMPORTANT: Enable EFA only on nodes with EFA devices attached.
+  # Otherwise, you'll encounter the "No devices found..." error. Restart the pod after attaching an EFA device, or use a node selector to prevent incompatible scheduling.
+  enable_aws_efa_k8s_device_plugin = var.enable_aws_efa_k8s_device_plugin
   aws_efa_k8s_device_plugin_helm_config = {
     values = [file("${path.module}/helm-values/aws-efa-k8s-device-plugin-values.yaml")]
   }
 
+  #---------------------------------------------------------------
+  # Karpenter Resources Add-on
+  #---------------------------------------------------------------
   enable_karpenter_resources = true
   karpenter_resources_helm_config = {
     g5-gpu-karpenter = {

ai-ml/jark-stack/terraform/install.sh

@@ -1,8 +1,5 @@
 #!/bin/bash
 
-read -p "Enter the region: " region
-export AWS_DEFAULT_REGION=$region
-
 # List of Terraform modules to apply in sequence
 targets=(
   "module.vpc"

ai-ml/jark-stack/terraform/main.tf

@@ -2,6 +2,13 @@ provider "aws" {
   region = local.region
 }
 
+# ECR always authenticates with `us-east-1` region
+# Docs -> https://docs.aws.amazon.com/AmazonECR/latest/public/public-registries.html
+provider "aws" {
+  alias  = "ecr"
+  region = "us-east-1"
+}
+
 provider "kubernetes" {
   host                   = module.eks.cluster_endpoint
   cluster_ca_certificate = base64decode(module.eks.cluster_certificate_authority_data)
@@ -27,10 +34,16 @@ data "aws_eks_cluster_auth" "this" {
   name = module.eks.cluster_name
 }
 
+data "aws_availability_zones" "available" {}
+
+data "aws_ecrpublic_authorization_token" "token" {
+  provider = aws.ecr
+}
+
 locals {
   name   = var.name
   region = var.region
-  azs    = ["${local.region}c", "${local.region}d"]
+  azs    = slice(data.aws_availability_zones.available.names, 0, 2)
   tags = {
     Blueprint  = local.name
     GithubRepo = "github.com/awslabs/data-on-eks"

ai-ml/jark-stack/terraform/variables.tf

@@ -38,3 +38,9 @@ variable "huggingface_token" {
   default     = "DUMMY_TOKEN_REPLACE_ME"
   sensitive   = true
 }
+
+variable "enable_aws_efa_k8s_device_plugin" {
+  description = "Enable AWS EFA K8s Device Plugin"
+  type        = bool
+  default     = false
+}