foxhunt/docs/deployment/cloud-deployment.md

Cloud Deployment Guide

Last Updated: 2025-10-22 Supported Platforms: AWS, GCP, Azure Prerequisites: Cloud CLI, Terraform 1.5+ Deployment Time: 45-60 minutes

---

AWS EKS Deployment

Prerequisites

# Install AWS CLI
curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
unzip awscliv2.zip && sudo ./aws/install

# Configure AWS credentials
aws configure
# AWS Access Key ID: YOUR_ACCESS_KEY
# AWS Secret Access Key: YOUR_SECRET_KEY
# Default region: us-east-1

# Install eksctl
curl --silent --location "https://github.com/weibaohui/eksctl/releases/latest/download/eksctl_$(uname -s)_amd64.tar.gz" | tar xz -C /tmp
sudo mv /tmp/eksctl /usr/local/bin

Create EKS Cluster

# Create cluster configuration
cat > foxhunt-eks-cluster.yaml <<EOF
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig

metadata:
  name: foxhunt-prod
  region: us-east-1
  version: "1.28"

vpc:
  cidr: 10.0.0.0/16
  nat:
    gateway: HighlyAvailable

nodeGroups:
  - name: app-nodes
    instanceType: c5.4xlarge
    desiredCapacity: 6
    minSize: 3
    maxSize: 12
    volumeSize: 100
    volumeType: gp3
    labels:
      workload-type: application
    tags:
      Environment: production
      Project: foxhunt

  - name: data-nodes
    instanceType: r5.2xlarge
    desiredCapacity: 3
    minSize: 2
    maxSize: 6
    volumeSize: 500
    volumeType: gp3
    labels:
      workload-type: data
    taints:
      - key: workload-type
        value: data
        effect: NoSchedule

  - name: gpu-nodes
    instanceType: g4dn.xlarge  # T4 GPU
    desiredCapacity: 1
    minSize: 0
    maxSize: 3
    volumeSize: 100
    labels:
      workload-type: gpu
      accelerator: nvidia-t4
    taints:
      - key: nvidia.com/gpu
        value: present
        effect: NoSchedule

addons:
  - name: vpc-cni
  - name: coredns
  - name: kube-proxy
  - name: aws-ebs-csi-driver

iam:
  withOIDC: true
  serviceAccounts:
  - metadata:
      name: ebs-csi-controller-sa
      namespace: kube-system
    wellKnownPolicies:
      ebsCSIController: true

cloudWatch:
  clusterLogging:
    enableTypes: ["api", "audit", "authenticator", "controllerManager", "scheduler"]
EOF

# Create cluster (30-40 minutes)
eksctl create cluster -f foxhunt-eks-cluster.yaml

# Verify cluster
kubectl get nodes
# NAME                          STATUS   ROLES    AGE   VERSION
# ip-10-0-1-100.ec2.internal    Ready    <none>   5m    v1.28.0
# ip-10-0-2-100.ec2.internal    Ready    <none>   5m    v1.28.0
# ip-10-0-3-100.ec2.internal    Ready    <none>   5m    v1.28.0

Configure RDS PostgreSQL

# Create RDS subnet group
aws rds create-db-subnet-group \
  --db-subnet-group-name foxhunt-db-subnet \
  --db-subnet-group-description "Foxhunt PostgreSQL subnet group" \
  --subnet-ids subnet-abc123 subnet-def456 subnet-ghi789

# Create RDS instance (TimescaleDB via custom AMI or Aurora PostgreSQL)
aws rds create-db-instance \
  --db-instance-identifier foxhunt-postgres \
  --db-instance-class db.r5.2xlarge \
  --engine postgres \
  --engine-version 15.4 \
  --master-username foxhunt \
  --master-user-password ${DB_PASSWORD} \
  --allocated-storage 500 \
  --storage-type gp3 \
  --storage-encrypted \
  --vpc-security-group-ids sg-abc123 \
  --db-subnet-group-name foxhunt-db-subnet \
  --backup-retention-period 30 \
  --preferred-backup-window "03:00-04:00" \
  --multi-az

# Wait for instance to be available
aws rds wait db-instance-available --db-instance-identifier foxhunt-postgres

# Get endpoint
aws rds describe-db-instances \
  --db-instance-identifier foxhunt-postgres \
  --query 'DBInstances[0].Endpoint.Address' \
  --output text
# foxhunt-postgres.abc123.us-east-1.rds.amazonaws.com

Configure ElastiCache Redis

# Create Redis cluster
aws elasticache create-replication-group \
  --replication-group-id foxhunt-redis \
  --replication-group-description "Foxhunt Redis cluster" \
  --engine redis \
  --cache-node-type cache.r5.large \
  --num-cache-clusters 3 \
  --automatic-failover-enabled \
  --multi-az-enabled \
  --cache-subnet-group-name foxhunt-cache-subnet \
  --security-group-ids sg-def456 \
  --at-rest-encryption-enabled \
  --transit-encryption-enabled \
  --auth-token ${REDIS_PASSWORD}

# Wait for cluster to be available
aws elasticache wait replication-group-available \
  --replication-group-id foxhunt-redis

# Get endpoint
aws elasticache describe-replication-groups \
  --replication-group-id foxhunt-redis \
  --query 'ReplicationGroups[0].NodeGroups[0].PrimaryEndpoint.Address' \
  --output text
# foxhunt-redis.abc123.cache.amazonaws.com

Deploy Application

# Update Helm values for AWS
cat > values-aws-prod.yaml <<EOF
global:
  environment: production
  imageRegistry: <AWS_ACCOUNT_ID>.dkr.ecr.us-east-1.amazonaws.com

postgresql:
  enabled: false  # Using RDS
  externalHost: foxhunt-postgres.abc123.us-east-1.rds.amazonaws.com
  externalPort: 5432

redis:
  enabled: false  # Using ElastiCache
  externalHost: foxhunt-redis.abc123.cache.amazonaws.com
  externalPort: 6379

ingress:
  enabled: true
  className: alb
  annotations:
    alb.ingress.kubernetes.io/scheme: internet-facing
    alb.ingress.kubernetes.io/target-type: ip
    alb.ingress.kubernetes.io/listen-ports: '[{"HTTPS":443}]'
    alb.ingress.kubernetes.io/certificate-arn: arn:aws:acm:us-east-1:123456789012:certificate/abc-123
EOF

# Deploy Helm chart
helm install foxhunt ./helm/foxhunt \
  --namespace foxhunt \
  --create-namespace \
  --values values-aws-prod.yaml \
  --timeout 15m \
  --wait

# Verify deployment
kubectl get pods -n foxhunt
kubectl get ingress -n foxhunt

---

GCP GKE Deployment

Create GKE Cluster

# Set project
gcloud config set project foxhunt-prod

# Create cluster
gcloud container clusters create foxhunt-prod \
  --zone us-central1-a \
  --num-nodes 6 \
  --machine-type n1-standard-8 \
  --disk-type pd-ssd \
  --disk-size 100 \
  --enable-autoscaling \
  --min-nodes 3 \
  --max-nodes 12 \
  --enable-stackdriver-kubernetes \
  --addons HorizontalPodAutoscaling,HttpLoadBalancing,GcePersistentDiskCsiDriver

# Add GPU node pool
gcloud container node-pools create gpu-pool \
  --cluster foxhunt-prod \
  --zone us-central1-a \
  --machine-type n1-standard-4 \
  --accelerator type=nvidia-tesla-t4,count=1 \
  --num-nodes 1 \
  --min-nodes 0 \
  --max-nodes 3 \
  --enable-autoscaling

# Get credentials
gcloud container clusters get-credentials foxhunt-prod --zone us-central1-a

Configure Cloud SQL PostgreSQL

# Create Cloud SQL instance
gcloud sql instances create foxhunt-postgres \
  --database-version=POSTGRES_15 \
  --tier=db-custom-8-32768 \
  --region=us-central1 \
  --storage-size=500 \
  --storage-type=SSD \
  --storage-auto-increase \
  --backup \
  --backup-start-time=03:00 \
  --maintenance-window-day=SUN \
  --maintenance-window-hour=04

# Set password
gcloud sql users set-password postgres \
  --instance=foxhunt-postgres \
  --password=${DB_PASSWORD}

# Get connection name
gcloud sql instances describe foxhunt-postgres \
  --format='value(connectionName)'
# foxhunt-prod:us-central1:foxhunt-postgres

---

Azure AKS Deployment

Create AKS Cluster

# Create resource group
az group create --name foxhunt-prod --location eastus

# Create AKS cluster
az aks create \
  --resource-group foxhunt-prod \
  --name foxhunt-prod-aks \
  --node-count 6 \
  --node-vm-size Standard_D8s_v3 \
  --enable-cluster-autoscaler \
  --min-count 3 \
  --max-count 12 \
  --enable-addons monitoring \
  --generate-ssh-keys

# Add GPU node pool
az aks nodepool add \
  --resource-group foxhunt-prod \
  --cluster-name foxhunt-prod-aks \
  --name gpupool \
  --node-count 1 \
  --node-vm-size Standard_NC4as_T4_v3 \
  --enable-cluster-autoscaler \
  --min-count 0 \
  --max-count 3

# Get credentials
az aks get-credentials --resource-group foxhunt-prod --name foxhunt-prod-aks

Configure Azure Database for PostgreSQL

# Create PostgreSQL server
az postgres flexible-server create \
  --resource-group foxhunt-prod \
  --name foxhunt-postgres \
  --location eastus \
  --admin-user foxhunt \
  --admin-password ${DB_PASSWORD} \
  --sku-name Standard_D8s_v3 \
  --tier GeneralPurpose \
  --storage-size 512 \
  --version 15 \
  --high-availability Enabled

# Get connection string
az postgres flexible-server show \
  --resource-group foxhunt-prod \
  --name foxhunt-postgres \
  --query "fullyQualifiedDomainName" \
  --output tsv
# foxhunt-postgres.postgres.database.azure.com

---

Cost Optimization

AWS Cost Estimates

Resource	Type	Monthly Cost
EKS Control Plane	1 cluster	$73
EC2 App Nodes	6x c5.4xlarge	$3,672
EC2 Data Nodes	3x r5.2xlarge	$1,814
EC2 GPU Nodes	1x g4dn.xlarge	$526
RDS PostgreSQL	db.r5.2xlarge Multi-AZ	$1,248
ElastiCache Redis	3x cache.r5.large	$621
Data Transfer	1TB/month	$90
Total		~$8,044/month

Cost Optimization Strategies

1. Reserved Instances: Save 40-60% on EC2/RDS costs
2. Spot Instances: Use for non-critical backtesting workloads
3. Auto-scaling: Scale down during off-hours (weekends, nights)
4. Storage Optimization: Use gp3 instead of io1 for general workloads
5. Right-sizing: Monitor actual usage and adjust instance types

---

End of Cloud Deployment Guide