Terraform (AWS)

The HEBBS Terraform module provisions an AWS EKS cluster pre-configured for running HEBBS, including networking, storage, and IAM configuration.

Prerequisites

Terraform >= 1.5
AWS CLI configured with appropriate credentials
kubectl installed locally
An AWS account with permissions to create VPC, EKS, EC2, EBS, and IAM resources

Module Source

module "hebbs" {
  source  = "github.com/hebbs-ai/terraform-aws-hebbs?ref=v0.1.0"

  cluster_name    = "hebbs-production"
  region          = "us-east-1"
  instance_type   = "m6i.xlarge"
  node_count      = 2
  ebs_volume_size = 100

  hebbs_version   = "0.1.0"
  enable_tls      = true
  enable_monitoring = true

  tags = {
    Environment = "production"
    Team        = "platform"
  }
}

Deployment Steps

1. Initialize

terraform init

2. Plan

terraform plan -out=tfplan

Review the plan to confirm the resources that will be created.

3. Apply

terraform apply tfplan

This typically takes 15–20 minutes to provision the EKS cluster and deploy HEBBS.

4. Configure kubectl

aws eks update-kubeconfig \
  --name hebbs-production \
  --region us-east-1

5. Verify

kubectl get pods -n hebbs
hebbs-cli status --address $(terraform output -raw hebbs_endpoint)

Module Variables

Variable	Default	Description
`cluster_name`	(required)	EKS cluster name
`region`	(required)	AWS region
`instance_type`	`m6i.large`	EC2 instance type for worker nodes
`node_count`	`2`	Number of worker nodes
`ebs_volume_size`	`50`	EBS volume size in GB for HEBBS data
`hebbs_version`	`latest`	HEBBS image tag
`enable_tls`	`false`	Enable TLS with ACM certificate
`enable_monitoring`	`false`	Deploy Prometheus + Grafana stack
`vpc_cidr`	`10.0.0.0/16`	VPC CIDR block
`tags`	`{}`	Tags to apply to all resources

Outputs

Output	Description
`hebbs_endpoint`	gRPC endpoint address (NLB DNS name)
`metrics_endpoint`	Prometheus metrics endpoint
`grafana_url`	Grafana dashboard URL (if monitoring enabled)
`cluster_name`	EKS cluster name
`kubeconfig_command`	Command to configure kubectl

Architecture

The module creates:

VPC with public and private subnets across 3 availability zones
EKS cluster with managed node group
EBS CSI driver for persistent volume provisioning (gp3)
Network Load Balancer for gRPC traffic
IAM roles for EKS, worker nodes, and EBS CSI
Helm release of the HEBBS chart with configured values

Cost Estimate

Component	Approximate Monthly Cost
EKS control plane	$73
2x m6i.xlarge nodes	$280
100 GB gp3 EBS	$8
Network Load Balancer	$16 + data transfer
Total	~$377/month

Destroying

terraform destroy

This removes all AWS resources. Data stored in EBS volumes is permanently deleted.