OSS Kubernetes di GCE dengan TPU, DRA untuk TPU, DRANET (OSS), dan Gemma 4

1. Ringkasan

Lab ini memperkenalkan cara membangun infrastruktur AI yang dikelola sendiri langsung di Google Compute Engine (GCE). Anda akan melakukan bootstrapping cluster Kubernetes yang tidak dikelola di virtual machine (beberapa di antaranya dengan TPU) menggunakan Terraform, kubeadm, dan mengonfigurasi Alokasi Resource Dinamis (DRA) Kubernetes menggunakan driver open source. Anda akan menggunakan hal berikut:

Untuk mengonfigurasi lingkungan, Anda akan men-deploy beberapa jaringan VPC independen yang masing-masing memiliki subnetnya sendiri. Hal ini memungkinkan Anda menyediakan instance VM dengan beberapa antarmuka jaringan (multi-NIC), sehingga memisahkan traffic pengelolaan dari traffic data TPU berkecepatan tinggi.

Selanjutnya, untuk mengaktifkan Alokasi Resource Dinamis (DRA) open source, Anda akan menginstal driver hardware Google TPU DRA dan driver jaringan DRANET. Kemudian, Anda akan mengonfigurasi DeviceClass Kubernetes dan menulis ResourceClaimTemplates untuk menangani penyediaan dinamis resource ini.

Terakhir, Anda akan men-deploy beban kerja benchmarking berperforma tinggi menggunakan Neper untuk memvalidasi jalur data jaringan Jumbo Frame antara node pekerja, diikuti dengan pengujian Python JAX untuk memvalidasi silikon TPU yang mendasarinya. Kemudian, Anda akan men-deploy vLLM untuk menyajikan model Gemma 4 mutakhir Google melalui Hugging Face menggunakan klaim DRA jaringan dan hardware yang terisolasi sepenuhnya.

Konfigurasi akan menggunakan kombinasi Terraform, gcloud, dan kubectl.

Di lab ini, Anda akan mempelajari cara melakukan tugas berikut:

  • Menyiapkan jaringan VPC
  • Men-deploy 3 node di GCE (1 node Standard dan 2 node TPU v6)
  • Bootstrap Kubernetes
  • Mengonfigurasi OSS DRANET dan DRA untuk TPU
  • Performa Tolok Ukur
  • Membuat DeviceClass dan ResourceClaimTemplate
  • Membuat tolok ukur performa jaringan dan hardware
  • Men-deploy Gemma 4: Menyajikan model di hardware TPU v6e menggunakan vLLM dan klaim DRA aktif
  • Menguji konektivitas ke LLM

Dalam lab ini, Anda akan membuat pola berikut.

Gambar 1.

b2f744fcb0c9b4df.jpeg

2. Penyiapan layanan Google Cloud

Penyiapan lingkungan mandiri

  1. Login ke Google Cloud Console dan buat project baru atau gunakan kembali project yang sudah ada. Jika belum memiliki akun Gmail atau Google Workspace, Anda harus membuatnya.

295004821bab6a87.png37d264871000675d.png96d86d3d5655cdbe.png

  • Project name adalah nama tampilan untuk peserta project ini. String ini adalah string karakter yang tidak digunakan oleh Google API. Anda dapat memperbaruinya kapan saja.
  • Project ID bersifat unik di semua project Google Cloud dan tidak dapat diubah (tidak dapat diubah setelah ditetapkan). Cloud Console otomatis membuat string unik; biasanya Anda tidak mementingkan kata-katanya. Di sebagian besar codelab, Anda harus merujuk Project ID-nya (umumnya diidentifikasi sebagai PROJECT_ID). Jika tidak suka dengan ID yang dibuat, Anda dapat membuat ID acak lainnya. Atau, Anda dapat mencobanya sendiri, dan lihat apakah ID tersebut tersedia. ID tidak dapat diubah setelah langkah ini dan tersedia selama durasi project.
  • Sebagai informasi, ada nilai ketiga, Project Number, yang digunakan oleh beberapa API. Pelajari lebih lanjut ketiga nilai ini di dokumentasi.
  1. Selanjutnya, Anda harus mengaktifkan penagihan di Konsol Cloud untuk menggunakan resource/API Cloud. Menjalankan operasi dalam codelab ini tidak akan memakan banyak biaya, bahkan mungkin tidak sama sekali. Guna mematikan resource agar tidak menimbulkan penagihan di luar tutorial ini, Anda dapat menghapus resource yang dibuat atau menghapus project-nya. Pengguna baru Google Cloud memenuhi syarat untuk mengikuti program Uji Coba Gratis senilai $300 USD.

Mulai Cloud Shell

Meskipun Google Cloud dapat dioperasikan dari jarak jauh menggunakan laptop Anda, dalam codelab ini, Anda akan menggunakan Google Cloud Shell, lingkungan command line yang berjalan di Cloud.

Dari Google Cloud Console, klik ikon Cloud Shell di toolbar kanan atas:

Mengaktifkan Cloud Shell

Hanya perlu waktu beberapa saat untuk penyediaan dan terhubung ke lingkungan. Jika sudah selesai, Anda akan melihat tampilan seperti ini:

Screenshot terminal Google Cloud Shell yang menunjukkan bahwa lingkungan telah terhubung

Mesin virtual ini berisi semua alat pengembangan yang Anda perlukan. Layanan ini menawarkan direktori beranda tetap sebesar 5 GB dan beroperasi di Google Cloud, sehingga sangat meningkatkan performa dan autentikasi jaringan. Semua pekerjaan Anda dalam codelab ini dapat dilakukan di browser. Anda tidak perlu menginstal apa pun.

3. Menyiapkan lingkungan dengan Terraform

Untuk mengerjakan lab ini, Anda memerlukan akses ke TPU. Versi persis yang digunakan adalah TPU v6e.

  • Anda harus mengikuti dokumen rencana TPU dan mengaktifkan kuota TPU untuk mendapatkan akses.
  • Gunakan region tempat Anda memiliki kuota TPU. Untuk mengetahui info selengkapnya, lihat dokumen ini " Memvalidasi ketersediaan TPU di GKE"
  • Kami menggunakan deployment kecil yang memerlukan (2) chip TPU v6e 4 (ct6e-standard-4t)yang akan menjadi slice 2x2 di satu region.
  • Token Hugging Face: Token Akses diperlukan untuk mendownload bobot model Gemma

Kita akan membuat tiga VPC kustom dengan aturan firewall dan subnet. Buka konsol cloud dan pilih project yang akan Anda gunakan.

  1. Buka Cloud Shell yang berada di bagian atas konsol Anda di sebelah kanan, pastikan Anda melihat project id yang benar di Cloud Shell, dan konfirmasi setiap perintah untuk mengizinkan akses. b51b80043d3bac90.png
  2. Buat folder bernama oss-kube-dra,, buka folder tersebut, lalu tambahkan beberapa variabel. p.s. Perbarui nilai variabel untuk "REGION", dan "ZONE" ke region dan zona sebenarnya, region default yang digunakan adalah "europe-west4" dan zona default yang digunakan adalah "europe-west4-a".
mkdir -p oss-kube-dra && cd oss-kube-dra
export PROJECT_ID=$(gcloud config get-value project)
export REGION="europe-west4" 
export ZONE="europe-west4-a" 
echo $PROJECT_ID
echo $REGION
echo $ZONE
  1. Sekarang, tambahkan beberapa file konfigurasi. Perintah ini akan membuat file terraform.tfvars , variables.tf, vpc.tf berikut.
cat << EOF > terraform.tfvars
project_id = "${PROJECT_ID}"
region     = "${REGION}"
zone       = "${ZONE}"
EOF

cat << 'EOF' > variables.tf
variable "project_id" {
  type        = string
  description = "The Google Cloud Project ID"
}

variable "region" {
  type        = string
  description = "The region to deploy the resources"
}

variable "zone" {
  type        = string
  description = "The specific zone for the VMs"
}

variable "control_plane_machine_type" {
  type        = string
  default     = "e2-standard-8"
  description = "Machine type for the Kubernetes control plane node"
}

variable "tpu_worker_machine_type" {
  type        = string
  default     = "ct6e-standard-4t"
  description = "The machine type for TPU workers (TPU v6e Trillium VM)"
}
EOF


cat << 'EOF' > vpc.tf
terraform {
  required_version = ">= 1.5.0"
  required_providers {
    google = {
      source  = "hashicorp/google"
      version = "~> 7.32.0"
    }
  }
}

provider "google" {
  project = var.project_id
  region  = var.region
}

# 1. Primary Management VPC and Subnet
resource "google_compute_network" "primary_vpc" {
  name                    = "oss-k8s-primary-vpc"
  auto_create_subnetworks = false
  mtu                     = 1460
}

resource "google_compute_subnetwork" "primary_subnet" {
  name          = "oss-k8s-primary-subnet"
  ip_cidr_range = "10.0.0.0/24"
  region        = var.region
  network       = google_compute_network.primary_vpc.id
}

# 2. Cloud NAT Router and NAT Gateway for Primary VPC (Outbound Access)
resource "google_compute_router" "router" {
  name    = "oss-k8s-router"
  network = google_compute_network.primary_vpc.id
  region  = var.region
}

resource "google_compute_router_nat" "nat" {
  name                               = "oss-k8s-nat"
  router                             = google_compute_router.router.name
  region                             = var.region
  nat_ip_allocate_option             = "AUTO_ONLY"
  source_subnetwork_ip_ranges_to_nat = "ALL_SUBNETWORKS_ALL_IP_RANGES"
}

# 3. Firewalls for Primary VPC
resource "google_compute_firewall" "allow_internal" {
  name    = "oss-k8s-primary-allow-internal"
  network = google_compute_network.primary_vpc.id

  allow {
    protocol = "tcp"
  }
  allow {
    protocol = "udp"
  }
  allow {
    protocol = "icmp"
  }

  source_ranges = ["10.0.0.0/24"]
}

resource "google_compute_firewall" "allow_iap" {
  name    = "oss-k8s-allow-iap-ssh"
  network = google_compute_network.primary_vpc.id

  allow {
    protocol = "tcp"
    ports    = ["22"]
  }

  source_ranges = ["35.235.240.0/20"]
}

# 4. Multi-NIC TPU Networks and Subnets (With Jumbo Frames MTU 8896)
resource "google_compute_network" "tpu_vpc" {
  count                   = 2
  name                    = "oss-tpu-vpc-${count.index + 1}"
  auto_create_subnetworks = false
  mtu                     = 8896
}

resource "google_compute_subnetwork" "tpu_subnet" {
  count         = 2
  name          = "oss-tpu-vpc-${count.index + 1}-subnet"
  ip_cidr_range = "10.${count.index + 1}0.0.0/24"
  region        = var.region
  network       = google_compute_network.tpu_vpc[count.index].id
}

resource "google_compute_firewall" "tpu_allow_internal" {
  count   = 2
  name    = "oss-tpu${count.index + 1}-allow-internal"
  network = google_compute_network.tpu_vpc[count.index].id

  allow {
    protocol = "tcp"
  }
  allow {
    protocol = "udp"
  }
  allow {
    protocol = "icmp"
  }

  source_ranges = ["10.${count.index + 1}0.0.0/24"]
}
EOF
  1. Pastikan Anda berada di direktori oss-kube-dra dan jalankan perintah berikut
    terraform init Menginisialisasi direktori kerja. Ini adalah langkah pertama dan akan mendownload penyedia yang diperlukan untuk konfigurasi tertentu.
    terraform plan -out membuat rencana eksekusi, yang menunjukkan tindakan yang akan dilakukan Terraform untuk men-deploy infrastruktur Anda. -out memungkinkan Anda menyimpan rencana eksekusi ke biner bernama. Anda dapat melihat apa yang akan terjadi tanpa melakukan perubahan apa pun.
    terraform apply menjalankan pembaruan.
terraform init 
terraform plan -out=tfplan
  1. Sekarang jalankan deployment setelah Anda menjalankan terraform apply, karena Anda menerapkan rencana eksekusi yang disimpan, rencana tersebut akan segera dijalankan tanpa meminta konfirmasi. (Proses ini dapat memakan waktu antara 5 -10 menit)
terraform apply tfplan
  1. Verifikasi penyiapan.
echo -e "\n=== Verifying VPC Networks ==="
gcloud compute networks list --filter="name~oss-.*" --project=$PROJECT_ID

echo -e "\n=== Verifying Subnetworks ==="
gcloud compute networks subnets list --filter="name~oss-.*" --project=$PROJECT_ID

echo -e "\n=== Verifying Firewall Rules ==="
gcloud compute firewall-rules list --filter="name~oss-.*" --project=$PROJECT_ID

echo -e "\n=== Verifying Cloud NAT ==="
gcloud compute routers nats list --router=oss-k8s-router --router-region=$REGION --project=$PROJECT_ID

Buat node VM Anda

Sekarang, Anda akan menentukan instance Compute Engine.

  1. Pastikan Anda berada di direktori oss-kube-dra dan jalankan perintah berikut di Cloud Shell untuk menulis file nodes.tf.
cat << 'EOF' > nodes.tf
# 1. K8s Control Plane VM (No TPU)
resource "google_compute_instance" "control_plane" {
  name         = "k8s-control-plane"
  machine_type = var.control_plane_machine_type
  zone         = var.zone

  boot_disk {
    initialize_params {
      image = "projects/ubuntu-os-cloud/global/images/family/ubuntu-2204-lts"
      size  = 100
    }
  }

  network_interface {
    network    = google_compute_network.primary_vpc.id
    subnetwork = google_compute_subnetwork.primary_subnet.id
    # No public IP block keeps this node private
  }

  service_account {
    scopes = ["cloud-platform"]
  }
}

# 2. TPU Worker VMs (Multi-NIC ct6e-standard-4t instances)
resource "google_compute_instance" "tpu_workers" {
  count        = 2
  name         = "k8s-tpu-worker-${count.index + 1}"
  machine_type = var.tpu_worker_machine_type
  zone         = var.zone

  boot_disk {
    initialize_params {
      image = "projects/ubuntu-os-accelerator-images/global/images/family/ubuntu-accel-2204-amd64-tpu-v5e-v5p-v6e"
      size  = 200
    }
  }

  scheduling {
    on_host_maintenance = "TERMINATE"
    provisioning_model  = "STANDARD"
  }

  # NIC 1: Management VPC Subnet
  network_interface {
    network    = google_compute_network.primary_vpc.id
    subnetwork = google_compute_subnetwork.primary_subnet.id
  }

  # NIC 2: TPU VPC 1 Subnet
  network_interface {
    network    = google_compute_network.tpu_vpc[0].id
    subnetwork = google_compute_subnetwork.tpu_subnet[0].id
  }

  # NIC 3: TPU VPC 2 Subnet
  network_interface {
    network    = google_compute_network.tpu_vpc[1].id
    subnetwork = google_compute_subnetwork.tpu_subnet[1].id
  }

  service_account {
    scopes = ["cloud-platform"]
  }

  
  lifecycle {
    ignore_changes = [
      boot_disk[0].initialize_params[0].image,
      guest_accelerator,
      metadata
    ]
  }
}
EOF
  1. Setelah konfigurasi baru ditulis, buat rencana baru dan terapkan untuk menyediakan instance Anda.
terraform plan -out=tfplan

terraform apply tfplan
  1. Verifikasi.
echo -e "\n=== Verifying Provisioned VM Instances ==="
gcloud compute instances list --filter="name~k8s-.*" --project=$PROJECT_ID


echo -e "\n=== Verifying Network Interfaces on Workers ==="
for i in 1 2; do
  echo -e "\n--- Interfaces for k8s-tpu-worker-${i} ---"
  gcloud compute instances describe k8s-tpu-worker-${i} \
      --zone=$ZONE \
      --project=$PROJECT_ID \
      --format="table(networkInterfaces[].network.basename(), networkInterfaces[].networkIP)"
done

4. Mem-bootstrap node kontrol cluster Kubernetes Anda

Di bagian ini, Anda akan terhubung dengan aman ke instance VM bidang kontrol yang baru dibuat, mengonfigurasi sistem operasi yang mendasarinya, menginstal paket Kubernetes dan runtime container, menginisialisasi cluster, serta men-deploy Calico CNI dengan isolasi traffic yang ketat ke jaringan pengelolaan.

  1. Hubungkan secara aman ke instance k8s-control-plane menggunakan tunnel Identity-Aware Proxy (IAP) GCE. Jalankan perintah berikut di terminal Cloud Shell:
gcloud compute ssh k8s-control-plane \
    --zone=$ZONE \
    --tunnel-through-iap
  1. Di VM k8s-control-plane, buat skrip bernama init-control-plane.sh untuk mengotomatiskan langkah-langkah penginstalan dan konfigurasi.
cat << 'CONTROL_PLANE_EOF' > init-control-plane.sh
#!/bin/bash
# Strict error handling: fail instantly if any command exits with a non-zero status
set -e

echo "=== 1. Neutralizing Background Updates & Preparing Base OS ==="
# Prevent unattended upgrades from locking apt or breaking network configuration mid-setup
sudo systemctl stop apt-daily.timer apt-daily-upgrade.timer || true
sudo systemctl disable apt-daily.timer apt-daily-upgrade.timer || true
sudo systemctl mask apt-daily.service apt-daily-upgrade.service || true

# Turn off swap (mandatory for Kubernetes)
sudo swapoff -a
sudo sed -i '/ swap / s/^\(.*\)$/#\1/g' /etc/fstab

# Load required kernel modules
cat << 'EOT' | sudo tee /etc/modules-load.d/k8s.conf
overlay
br_netfilter
EOT

sudo modprobe overlay
sudo modprobe br_netfilter

# Configure sysctl requirements for Kubernetes bridging
cat << 'EOT' | sudo tee /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-iptables  = 1
net.bridge.bridge-nf-call-ip6tables = 1
net.ipv4.ip_forward                 = 1
EOT
sudo sysctl --system

echo "=== 2. Installing Container Runtime (Containerd) ==="
sudo apt-get update
sudo apt-get install -y ca-certificates curl gnupg bash-completion

sudo install -m 0755 -d /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor --yes -o /etc/apt/keyrings/docker.gpg
sudo chmod a+r /etc/apt/keyrings/docker.gpg
echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu $(. /etc/os-release && echo $VERSION_CODENAME) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

sudo apt-get update
sudo apt-get install -y docker-ce docker-ce-cli containerd.io

echo "=== 3. Configuring Containerd with Systemd Cgroups ==="
sudo mkdir -p /etc/containerd
containerd config default | sudo tee /etc/containerd/config.toml >/dev/null
sudo sed -i 's/SystemdCgroup = false/SystemdCgroup = true/' /etc/containerd/config.toml

sudo systemctl daemon-reload
sudo systemctl restart containerd
sudo systemctl enable containerd

# Validation Step: Verify runtime engine health
if ! systemctl is-active --quiet containerd; then
    echo "❌ ERROR: Containerd failed to start properly."
    exit 1
fi
echo "✅ Containerd runtime is active and healthy."

echo "=== 4. Installing Kubernetes 1.36 Binaries ==="
curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.36/deb/Release.key | sudo gpg --dearmor --yes -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg
echo 'deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] https://pkgs.k8s.io/core:/stable:/v1.36/deb/ /' | sudo tee /etc/apt/sources.list.d/kubernetes.list

sudo apt-get update
sudo apt-get install -y kubelet kubeadm kubectl
sudo apt-mark hold kubelet kubeadm kubectl

# Configure Autocomplete and Aliases system-wide
kubectl completion bash | sudo tee /etc/bash_completion.d/kubectl > /dev/null
kubeadm completion bash | sudo tee /etc/bash_completion.d/kubeadm > /dev/null
if ! grep -q 'alias k=kubectl' ~/.bashrc; then
  echo 'alias k=kubectl' >> ~/.bashrc
  echo 'complete -o default -F __start_kubectl k' >> ~/.bashrc
fi

echo "=== 5. Initializing Control Plane Engine ==="
sudo kubeadm init --pod-network-cidr=192.168.0.0/16

echo "=== 6. Configuring Administrative Cluster Credentials ==="
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config

# Validation Step: Verify API Server local responsiveness
echo "Waiting for local API server context..."
until kubectl cluster-info &>/dev/null; do
    sleep 2
done
echo "✅ Kubernetes API server is responding locally."

echo "=== 7. Deploying Calico Network Operator ==="
kubectl create -f https://raw.githubusercontent.com/projectcalico/calico/v3.27.3/manifests/tigera-operator.yaml

# Validation Step: Ensure Tigera Operator CRD is fully available before applying configuration
echo "Waiting for Tigera Installation CRD to register on the API server..."
kubectl wait --for=condition=established crd/installations.operator.tigera.io --timeout=60s

echo "=== 8. Deploying Calico Custom Resources (Subnet Interlock Locked to 10.0.0.0/24) ==="
cat << 'CALICO_EOF' > custom-calico.yaml
apiVersion: operator.tigera.io/v1
kind: Installation
metadata:
  name: default
spec:
  calicoNetwork:
    nodeAddressAutodetectionV4:
      cidrs:
        - "10.0.0.0/24"
    ipPools:
    - blockSize: 26
      cidr: 192.168.0.0/16
      encapsulation: VXLANCrossSubnet
      natOutgoing: Enabled
      nodeSelector: all()
CALICO_EOF
kubectl apply -f custom-calico.yaml

# Validation Step: Confirm Calico daemon configurations are processing
echo "Waiting 10 seconds for Calico system namespaces to initialize..."
sleep 10
echo "Current Calico workload deployment status:"
kubectl get pods -n calico-system

echo "=== 9. Exporting Worker Cluster Join Token ==="
sudo kubeadm token create --print-join-command > ~/join.sh
chmod +x ~/join.sh

echo "--------------------------------------------------------"
echo "✅ CONTROL PLANE BOOTSTRAP COMPLETE!"
echo "Your cluster join command for the TPU workers is saved below:"
echo "--------------------------------------------------------"
cat ~/join.sh
CONTROL_PLANE_EOF
  1. Jalankan skrip.
chmod +x init-control-plane.sh
./init-control-plane.sh
  1. Setelah selesai, verifikasi. Semua fitur akan aktif dalam beberapa menit.
kubectl get nodes
kubectl get pods -A

Anda akan melihat sesuatu yang mirip dengan ini

NAME                STATUS   ROLES           AGE     VERSION
k8s-control-plane   Ready    control-plane   6m50s   v1.36.2
NAMESPACE         NAME                                        READY   STATUS    RESTARTS   AGE
calico-system     calico-kube-controllers-5578ff64dd-87vp2    1/1     Running   0          6m33s
calico-system     calico-node-fxzpp                           1/1     Running   0          6m33s
calico-system     calico-typha-785cbc858-rv4nz                1/1     Running   0          6m33s
calico-system     csi-node-driver-wlrhx                       2/2     Running   0          6m33s
kube-system       coredns-589f44dc88-pqfrl                    1/1     Running   0          6m42s
kube-system       coredns-589f44dc88-sdwmj                    1/1     Running   0          6m42s
kube-system       etcd-k8s-control-plane                      1/1     Running   0          6m47s
kube-system       kube-apiserver-k8s-control-plane            1/1     Running   0          6m47s
kube-system       kube-controller-manager-k8s-control-plane   1/1     Running   0          6m47s
kube-system       kube-proxy-jnm2p                            1/1     Running   0          6m42s
kube-system       kube-scheduler-k8s-control-plane            1/1     Running   0          6m47s
tigera-operator   tigera-operator-6bc8d879b5-w5mrq            1/1     Running   0          6m42s
  1. Keluar dari koneksi ssh untuk kembali ke Cloud Shell
exit

5. Tambahkan node pekerja TPU

Anda akan menjalankan skrip dari Cloud Shell yang terhubung dengan aman ke VM bidang kontrol, mengambil token gabungan cluster, serta mengonfigurasi dan mendaftarkan node worker TPU ke cluster secara bersamaan.

  1. Jalankan perintah berikut di Cloud Shell untuk menulis skrip orkestrasi:
cat << 'WORKER_BOOTSTRAP_EOF' > bootstrap-workers.sh
#!/bin/bash
# Strict error handling: fail instantly if any command exits with a non-zero status
set -e

# Fetch the join command safely from the control plane
echo "Fetching join command from Control Plane..."
JOIN_CMD=$(gcloud compute ssh k8s-control-plane --zone=$ZONE --tunnel-through-iap --command="cat ~/join.sh" 2>/dev/null)

if [ -z "$JOIN_CMD" ]; then
    echo "❌ ERROR: Failed to retrieve the join command. Ensure the control plane is reachable."
    exit 1
fi

echo "✅ Successfully retrieved join command."

# Create the setup script locally to be copied to the workers
cat << 'WORKER_INIT_EOF' > init-worker.sh
#!/bin/bash
set -e

echo "=== 1. Neutralizing Background Updates & Setting Non-Interactive Mode ==="
export DEBIAN_FRONTEND=noninteractive
sudo sed -i "s/#\$nrconf{restart} = 'i';/\$nrconf{restart} = 'a';/g" /etc/needrestart/needrestart.conf 2>/dev/null || true

# Prevent unattended upgrades from tearing down network interfaces mid-setup
sudo systemctl stop apt-daily.timer apt-daily-upgrade.timer || true
sudo systemctl disable apt-daily.timer apt-daily-upgrade.timer || true
sudo systemctl mask apt-daily.service apt-daily-upgrade.service || true

echo "=== 2. Base OS Prep ==="
# Disable swap
sudo swapoff -a
sudo sed -i '/ swap / s/^\(.*\)$/#\1/g' /etc/fstab

# Load required kernel modules
cat << 'EOT' | sudo tee /etc/modules-load.d/k8s.conf
overlay
br_netfilter
EOT
sudo modprobe overlay
sudo modprobe br_netfilter

# Configure bridging and IP forwarding sysctls
cat << 'EOT' | sudo tee /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-iptables  = 1
net.bridge.bridge-nf-call-ip6tables = 1
net.ipv4.ip_forward                 = 1
EOT
sudo sysctl --system

echo "=== 3. Installing Containerd (CRI-Only) ==="
sudo apt-get update && sudo apt-get install -yq ca-certificates curl gnupg bash-completion
sudo install -m 0755 -d /etc/apt/keyrings
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor --yes -o /etc/apt/keyrings/docker.gpg
sudo chmod a+r /etc/apt/keyrings/docker.gpg
echo "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] https://download.docker.com/linux/ubuntu $(. /etc/os-release && echo $VERSION_CODENAME) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null

# Install only containerd to avoid unnecessary Docker CE overhead
sudo apt-get update && sudo apt-get install -yq containerd.io
sudo mkdir -p /etc/containerd
containerd config default | sudo tee /etc/containerd/config.toml >/dev/null
sudo sed -i 's/SystemdCgroup = false/SystemdCgroup = true/' /etc/containerd/config.toml
sudo systemctl daemon-reload
sudo systemctl restart containerd
sudo systemctl enable containerd

# Validation: Check containerd status
if ! systemctl is-active --quiet containerd; then
    echo "❌ ERROR: Containerd failed to start."
    exit 1
fi

echo "=== 4. Installing Kubernetes 1.36 Binaries ==="
curl -fsSL https://pkgs.k8s.io/core:/stable:/v1.36/deb/Release.key | sudo gpg --dearmor --yes -o /etc/apt/keyrings/kubernetes-apt-keyring.gpg
echo 'deb [signed-by=/etc/apt/keyrings/kubernetes-apt-keyring.gpg] https://pkgs.k8s.io/core:/stable:/v1.36/deb/ /' | sudo tee /etc/apt/sources.list.d/kubernetes.list
sudo apt-get update && sudo apt-get install -yq kubelet kubeadm kubectl
sudo apt-mark hold kubelet kubeadm kubectl
WORKER_INIT_EOF

# Append the actual join command to the script
echo "echo \"=== 5. Joining Cluster ===\"" >> init-worker.sh
echo "sudo $JOIN_CMD" >> init-worker.sh

# Push and run on both Workers concurrently
echo "Starting concurrent bootstrap on both workers..."

(
    echo "[Worker 1] Copying script..."
    gcloud compute scp init-worker.sh k8s-tpu-worker-1:~ --zone=$ZONE --tunnel-through-iap --quiet
    echo "[Worker 1] Executing script..."
    gcloud compute ssh k8s-tpu-worker-1 --zone=$ZONE --tunnel-through-iap --command="bash ~/init-worker.sh"
    echo "✅ [Worker 1] Bootstrap and Join complete!"
) &

(
    echo "[Worker 2] Copying script..."
    gcloud compute scp init-worker.sh k8s-tpu-worker-2:~ --zone=$ZONE --tunnel-through-iap --quiet
    echo "[Worker 2] Executing script..."
    gcloud compute ssh k8s-tpu-worker-2 --zone=$ZONE --tunnel-through-iap --command="bash ~/init-worker.sh"
    echo "✅ [Worker 2] Bootstrap and Join complete!"
) &

# Wait for both background processes to finish
wait

echo "--------------------------------------------------------"
echo "✅ BOTH WORKERS HAVE FINISHED PROCESSING"
echo "--------------------------------------------------------"

# Final Validation Check from Control Plane
echo "Verifying cluster node status..."
sleep 5 # Give kubelet a moment to register the nodes
gcloud compute ssh k8s-control-plane --zone=$ZONE --tunnel-through-iap --command="kubectl get nodes -o wide"
WORKER_BOOTSTRAP_EOF
  1. Jalankan Penyiapan Pekerja. (Proses ini menjalankan kedua penginstalan secara bersamaan di latar belakang dan memerlukan waktu sekitar 3 hingga 5 menit untuk diselesaikan).
chmod +x bootstrap-workers.sh
./bootstrap-workers.sh

Anda akan melihat sesuatu yang serupa saat semua node ditambahkan ke cluster

To increase the performance of the tunnel, consider installing NumPy. For instructions,
please see https://cloud.google.com/iap/docs/using-tcp-forwarding#increasing_the_tcp_upload_bandwidth

NAME                STATUS     ROLES           AGE   VERSION   INTERNAL-IP   EXTERNAL-IP   OS-IMAGE             KERNEL-VERSION           CONTAINER-RUNTIME
k8s-control-plane   Ready      control-plane   25m   v1.36.2   10.0.0.2      <none>        Ubuntu 22.04.5 LTS   6.8.0-1064-gcp (amd64)   containerd://2.2.6
k8s-tpu-worker-1    NotReady   <none>          10s   v1.36.2   10.0.0.3      <none>        Ubuntu 22.04.5 LTS   6.8.0-1064-gcp (amd64)   containerd://2.2.6
k8s-tpu-worker-2    Ready      <none>          27s   v1.36.2   10.0.0.4      <none>        Ubuntu 22.04.5 LTS   6.8.0-1064-gcp (amd64)   containerd://2.2.6

6. Men-deploy Driver TPU DRA OSS

Di bagian ini, Anda akan kembali ke bidang kontrol, memberi label pada node pekerja TPU dengan detail topologi akselerator spesifiknya, dan menginstal driver DRA TPU Google open source menggunakan Helm. Driver ini bertanggung jawab untuk menemukan chip TPU v6e fisik dan memetakannya secara native ke Kubernetes API.

  1. Hubungkan kembali ke VM k8s-control-plane dari Cloud Shell dengan aman.
gcloud compute ssh k8s-control-plane \
    --zone=$ZONE \
    --tunnel-through-iap
  1. Jalankan perintah ini di dalam sesi SSH k8s-control-plane Anda. Beri Label pada Node dengan kumpulan label lengkap (termasuk kunci jumlah chip yang tepat)
kubectl label node k8s-tpu-worker-1 \
  cloud.google.com/gke-tpu-accelerator=tpu-v6e-slice \
  cloud.google.com/gke-tpu-topology=2x2 \
  cloud.google.com/gke-tpu-dra-driver=true \
  cloud.google.com/gke-accelerator-count=4 \
  cloud.google.com/gke-tpu-count=4 \
  --overwrite

kubectl label node k8s-tpu-worker-2 \
  cloud.google.com/gke-tpu-accelerator=tpu-v6e-slice \
  cloud.google.com/gke-tpu-topology=2x2 \
  cloud.google.com/gke-tpu-dra-driver=true \
  cloud.google.com/gke-accelerator-count=4 \
  cloud.google.com/gke-tpu-count=4 \
  --overwrite
  1. Meng-clone dan menginstal driver TPU DRA dengan Helm
curl -fsSL https://raw.githubusercontent.com/helm/helm/main/scripts/get-helm-3 | bash 

git clone https://github.com/kubernetes-sigs/dra-driver-google-tpu.git ~/dra-driver-google-tpu || true
cd ~/dra-driver-google-tpu
rm -f *.pack *.tgz

helm install dra-driver-google-tpu ./deployments/helm/dra-driver-google-tpu \
  -n dra-driver-google-tpu \
  --create-namespace \
  --set 'kubeletPlugin.env[0].name=NODE_NAME' \
  --set 'kubeletPlugin.env[0].valueFrom.fieldRef.fieldPath=spec.nodeName'

cd ~
  1. Memvalidasi penyiapan driver TPU DRA
# Verify driver daemonset status (Pods should show as Running and Ready)
kubectl get pods -n dra-driver-google-tpu -o wide

# Verify TPU ResourceSlices are successfully published to the API server
kubectl get resourceslices

# Safely parse the ResourceSlices to show the Node Name and the number of TPU chips registered
kubectl get resourceslices -o json | jq -r '.items[] | select(.spec.driver=="tpu.google.com") | "Node: \(.spec.nodeName) | TPUs Registered: \(.spec.devices | length)"'

# Inspect driver logs to confirm the TPU hardware was initialized successfully
kubectl logs -n dra-driver-google-tpu -l app.kubernetes.io/name=dra-driver-google-tpu -c tpu-dra-plugin --tail=20

7. Men-deploy DRANET Open Source & Kelas Perangkat

Di bagian ini, Anda akan kembali ke bidang kontrol, menginstal driver DRANET open source, menerapkan patch filter kustom untuk mengecualikan antarmuka virtual, serta membuat DeviceClass dan ResourceClaimTemplate Kubernetes dengan awalan jaringan OSS yang cocok.

  1. Hubungkan kembali ke VM k8s-control-plane dari Cloud Shell dengan aman. Jika sudah terhubung, lewati.
gcloud compute ssh k8s-control-plane \
    --zone=$ZONE \
    --tunnel-through-iap
  1. Jalankan perintah ini di dalam sesi k8s-control-plane SSH Anda
# Install the core components and patch
kubectl apply -f https://raw.githubusercontent.com/kubernetes-sigs/dranet/refs/heads/main/install.yaml


kubectl patch daemonset dranet -n kube-system --type='json' -p='[ { "op": "add", "path": "/spec/template/spec/containers/0/args/-", "value": "-filter=!(\"dra.net/type\" in attributes) || (attributes[\"dra.net/type\"].StringValue != \"veth\" && attributes[\"dra.net/type\"].StringValue != \"vxlan\" && attributes[\"dra.net/type\"].StringValue != \"bridge\")" } ]'

# Monitor rollout readiness
kubectl rollout status daemonset/dranet -n kube-system

# Verify running components and permissions
kubectl get pods -n kube-system -l app=dranet -o wide
kubectl get clusterrole,clusterrolebinding,sa dranet -n kube-system

# Interrogate logs for driver binding confirmation
kubectl logs -n kube-system -l app=dranet --tail=20
  1. Terapkan DeviceClass dan ResourceClaimTemplate
# Apply DRANET DeviceClass and BOTH ResourceClaimTemplates (Network + Hardware)
cat << 'EOF' | kubectl apply -f -
apiVersion: resource.k8s.io/v1
kind: DeviceClass
metadata:
  name: dranet
spec:
  selectors:
    - cel:
        expression: device.driver == "dra.net"
---
apiVersion: resource.k8s.io/v1
kind: ResourceClaimTemplate
metadata:
  name: tpu-net-interfaces
  namespace: default
spec:
  spec:
    devices:
      requests:
      - name: tpu-net-interface
        exactly:
          deviceClassName: dranet
          count: 2
          selectors:
          - cel:
              expression: device.attributes["gce.dra.net"].networkName.startsWith("oss-tpu-vpc")
      config:
      - opaque:
          driver: dra.net
          parameters:
            interface:
              mtu: 8896
              gsoMaxSize: 65536
              groMaxSize: 65536
              gsoIPv4MaxSize: 65536
              groIPv4MaxSize: 65536
              disableEbpfPrograms: true
---
apiVersion: resource.k8s.io/v1
kind: ResourceClaimTemplate
metadata:
  name: tpu-device-template
  namespace: default
spec:
  spec:
    devices:
      requests:
      - name: tpu-devices
        exactly:
          deviceClassName: tpu.google.com
          allocationMode: ExactCount
          count: 4
EOF
  1. Pastikan template dan class Anda terdaftar dengan benar di Kubernetes API.
# Verify ResourceSlices exist and are actively serving both drivers
kubectl get resourceslices -o custom-columns=NAME:.metadata.name,NODE:.spec.nodeName,DRIVER:.spec.driver | grep -E "dra.net|tpu.google.com"

# Verify the DRANET daemonset pods are Running across all nodes
kubectl get pods -n kube-system -l app=dranet -o wide
  1. Deploy StatefulSet Parallel Neper.
cat << 'EOF' | kubectl apply -f -
---
apiVersion: v1
kind: Service
metadata:
  name: neper
spec:
  clusterIP: None
  selector:
    app: neper
---
apiVersion: apps/v1
kind: StatefulSet
metadata:
  name: neper
spec:
  selector:
    matchLabels:
      app: neper
  serviceName: neper
  replicas: 2
  template:
    metadata:
      labels:
        app: neper
    spec:
      initContainers:
      - name: "network-optimization-sysctls"
        image: "busybox"
        securityContext:
          privileged: true
        command:
        - sh
        - -c
        - |
          echo 5000 > /proc/sys/net/ipv4/tcp_rto_min_us
          echo 1 > /proc/sys/net/ipv4/tcp_no_metrics_save
          echo 0 > /proc/sys/net/ipv4/tcp_slow_start_after_idle
          echo 131072 > /proc/sys/net/core/optmem_max
          echo "4096 41943040 314572800" > /proc/sys/net/ipv4/tcp_rmem          
      containers:
      - name: neper
        image: ubuntu:22.04
        command:
        - /bin/bash
        - -c
        - |
          apt-get update && apt-get install -y iproute2 build-essential git jq python3-pip &&
          git clone https://github.com/google/neper.git /tmp/neper &&
          cd /tmp/neper && make &&
          cp tcp_stream /usr/local/bin/ &&
          sleep infinity
        securityContext:
          privileged: true
        resources:
          requests:
            cpu: "170"
            memory: "650Gi"
          limits:
            cpu: "170"
            memory: "650Gi"
          claims:
          - name: tpu-net-claim
          - name: tpu-hardware-claim
      resourceClaims:
      - name: tpu-net-claim
        resourceClaimTemplateName: tpu-net-interfaces
      - name: tpu-hardware-claim
        resourceClaimTemplateName: tpu-device-template
EOF
  1. Pemeriksaan validasi
echo -e "\n=== Verifying StatefulSet Pod Status ==="
kubectl get pods -l app=neper -o wide

echo -e "\n=== Verifying Dynamic Resource Claims (DRCs) ==="
kubectl get resourceclaims

echo -e "\n=== Inspecting Device Claim Allocation ==="
# Using a safer JSONPath query to extract the allocated drivers and devices
kubectl get resourceclaims -o json | jq -r '.items[] | "Claim: \(.metadata.name) | Driver: \(.status.allocation.devices.results[0].driver // "Pending")"'

8. Jalankan pengujian

Menjalankan rangkaian validasi hardware dan tolok ukur antarmuka ganda.

Fase 1 (Tolok Ukur Jaringan): Fase ini menunggu kedua pod Neper (neper-0 dan neper-1) mengompilasi dependensi, mengekstrak alamat IP multi-NIC non-default yang terikat melalui DRANET, meluncurkan server tcp_stream serentak di neper-1, menghasilkan beban throughput tinggi dari neper-0, dan mengurai throughput gabungan dalam Gigabit per detik (Gbps).

Fase 2 (Validasi Hardware): Fase ini menginstal Google JAX di dalam neper-0 dan menjalankan perkalian matriks (5000x5000) langsung di chip TPU yang dipetakan melalui VFIO untuk mengonfirmasi status operasional silikon.

  1. Jalankan perintah berikut di k8s-control-plane untuk menulis run_dual_neper_test.sh
cat << 'EOF' > run_dual_neper_test.sh
#!/bin/bash
set -e

SERVER_POD="neper-1"
CLIENT_POD="neper-0"

echo "================================================="
echo " PHASE 1: DUAL-INTERFACE HIGH-SPEED NETWORK TEST"
echo "================================================="
echo "=== Waiting for Pods to be Ready ==="
kubectl wait --for=condition=ready pod/$CLIENT_POD pod/$SERVER_POD --timeout=300s

echo "=== Waiting for neper compilation to finish inside Pods ==="
for POD in $SERVER_POD $CLIENT_POD; do
  until kubectl exec $POD -c neper -- sh -c 'command -v jq >/dev/null 2>&1 && command -v tcp_stream >/dev/null 2>&1'; do
    sleep 5
  done
done

echo ""
echo "=== Step 1: Extract Target IPs from $SERVER_POD ==="
# Using jq to parse the network interfaces directly from Linux JSON output
IFACE1=$(kubectl exec $SERVER_POD -c neper -- sh -c "ip -j -4 addr show | jq -r '.[] | select(.ifname != \"lo\" and .ifname != \"eth0\") | .ifname' | sed -n '1p'")
IFACE2=$(kubectl exec $SERVER_POD -c neper -- sh -c "ip -j -4 addr show | jq -r '.[] | select(.ifname != \"lo\" and .ifname != \"eth0\") | .ifname' | sed -n '2p'")

IP1=$(kubectl exec $SERVER_POD -c neper -- sh -c "ip -j -4 addr show | jq -r '.[] | select(.ifname != \"lo\" and .ifname != \"eth0\") | .addr_info[0].local' | sed -n '1p'")
IP2=$(kubectl exec $SERVER_POD -c neper -- sh -c "ip -j -4 addr show | jq -r '.[] | select(.ifname != \"lo\" and .ifname != \"eth0\") | .addr_info[0].local' | sed -n '2p'")

echo "   📍 Target IP 1 ($IFACE1): $IP1"
echo "   📍 Target IP 2 ($IFACE2): $IP2"

echo ""
echo "=== Step 2: Initialize TCP Servers on $SERVER_POD ==="
kubectl exec $SERVER_POD -c neper -- sh -c '
for i in 0 1; do
  nohup tcp_stream -C$((52279 + i)) --port=$((38339 + i)) --skip-rx-copy -rw -Z -B16384 \
    --test-length=60 --suicide-length=120 -F100 --num-threads=16 --num-flows=32 -D0 \
    --logtostderr > test${i}.log 2>&1 &
done
'
sleep 3

echo "=== Step 3: Generate Concurrent High-Throughput Load from $CLIENT_POD ==="
echo "Blasting Traffic via Interface 1 -> $IP1 ..."
kubectl exec $CLIENT_POD -c neper -- sh -c "nohup tcp_stream -C52279 --port=38339 --skip-rx-copy -rw -Z -B16384 \
  --test-length=60 --suicide-length=70 -F100 --num-threads=16 --num-flows=32 \
  --client -H $IP1 -D0 --logtostderr > test0.log 2>&1 &"

echo "Blasting Traffic via Interface 2 -> $IP2 ..."
kubectl exec $CLIENT_POD -c neper -- sh -c "nohup tcp_stream -C52280 --port=38340 --skip-rx-copy -rw -Z -B16384 \
  --test-length=60 --suicide-length=70 -F100 --num-threads=16 --num-flows=32 \
  --client -H $IP2 -D0 --logtostderr > test1.log 2>&1 &"

echo ""
echo "=== Testing in progress... Waiting 65 seconds for test completion ==="
sleep 65

echo ""
echo "=== Step 4: Evaluate Throughput Metrics ==="
RAW_BPS1=$(kubectl exec $CLIENT_POD -c neper -- grep -a "remote_throughput=" test0.log | cut -d= -f2 | tr -d '\r' || echo "0")
RAW_BPS2=$(kubectl exec $CLIENT_POD -c neper -- grep -a "remote_throughput=" test1.log | cut -d= -f2 | tr -d '\r' || echo "0")

GBPS1=$(awk -v bps="$RAW_BPS1" 'BEGIN { printf "%.2f", bps / 1000000000 }')
GBPS2=$(awk -v bps="$RAW_BPS2" 'BEGIN { printf "%.2f", bps / 1000000000 }')
TOTAL=$(awk -v b1="$RAW_BPS1" -v b2="$RAW_BPS2" 'BEGIN { printf "%.2f", (b1 + b2) / 1000000000 }')

echo "📊 --- NETWORK RESULTS ---"
echo "Interface 1 ($IFACE1) : ${GBPS1} Gbps"
echo "Interface 2 ($IFACE2) : ${GBPS2} Gbps"
echo "🔥 TOTAL AGGREGATE  : ${TOTAL} Gbps"
echo "--------------------------"


echo ""
echo "================================================="
echo " PHASE 2: TPU HARDWARE VALIDATION TEST"
echo "================================================="
echo "⏳ Installing Python and Google JAX on $CLIENT_POD (Takes ~1 minute)..."
kubectl exec $CLIENT_POD -c neper -- bash -c "apt-get update > /dev/null 2>&1 && apt-get install -y python3-pip > /dev/null 2>&1 && pip3 install jax[tpu] -f https://storage.googleapis.com/jax-releases/libtpu_releases.html > /dev/null 2>&1"

echo "🧠 Running matrix math directly on the TPU chips..."
kubectl exec $CLIENT_POD -c neper -- python3 -c "
import jax
import jax.numpy as jnp
print(f'✅ TPU Hardware Detected: {jax.device_count()} chips mapped via vfio')
print('🚀 Executing 5000x5000 Matrix Multiplication on TPU silicon...')
x = jnp.ones((5000, 5000))
y = jnp.dot(x, x)
print('✅ Success! The TPU driver is fully operational and executing math.')
"
EOF

chmod +x run_dual_neper_test.sh
  1. Jalankan pengujian. Proses ini akan memerlukan waktu 2 menit.
./run_dual_neper_test.sh

Setelah selesai, output terminal Anda akan menampilkan metrik jaringan berkecepatan tinggi yang divalidasi dan eksekusi matematika matriks TPU

=== Step 4: Evaluate Throughput Metrics ===
📊 --- NETWORK RESULTS ---
Interface 1 (ens9)  : 157.51 Gbps
Interface 2 (ens10) : 167.04 Gbps
🔥 TOTAL AGGREGATE  : 324.55 Gbps
--------------------------

=================================================
 PHASE 2: TPU HARDWARE VALIDATION TEST
=================================================
⏳ Installing Python and Google JAX on neper-0 (Takes ~1 minute)...
🧠 Running matrix math directly on the TPU chips...
✅ TPU Hardware Detected: 4 chips mapped via vfio
🚀 Executing 5000x5000 Matrix Multiplication on TPU silicon...
✅ Success! The TPU driver is fully operational and executing math.

9. Men-deploy Gemma 4 di cluster Anda

Di bagian ini, Anda akan mengonfigurasi kredensial Hugging Face API yang aman sebagai secret Kubernetes, men-deploy mesin inferensi vLLM menggunakan klaim jaringan dan hardware Alokasi Resource Dinamis (DRA), serta menjalankan kueri pengujian end-to-end terhadap model Gemma 4 Google.

Pastikan Anda login ke sesi SSH aman di k8s-control-plane:

  1. Hubungkan kembali dengan aman ke VM bidang kontrol dari Cloud Shell. Jika Anda sudah terhubung, lewati langkah ini.
gcloud compute ssh k8s-control-plane \
    --zone=$ZONE \
    --tunnel-through-iap
  1. Membersihkan deployment sebelumnya
# 1. Delete the StatefulSet to stop the benchmarking pods
kubectl delete statefulset neper

# 2. Wait for the pods to terminate fully and release the claims
kubectl wait --for=delete pod/neper-0 pod/neper-1 --timeout=60s
  1. Simpan Token Akses Hugging Face Anda. Ganti <YOUR_ACTUAL_HUGGING_FACE_TOKEN> dengan token Anda.
export HF_TOKEN="<YOUR_ACTUAL_HUGGING_FACE_TOKEN>"
  1. Membuat secret
kubectl create secret generic hf-token --from-literal=token="${HF_TOKEN}"
  1. Manifes ini menjadwalkan satu replika vLLM yang berjalan di VM TPU mentah 4 chip. Hal ini memanfaatkan standar DRA Kubernetes untuk memasang klaim jaringan kustom (tpu-net-claim) dan klaim hardware (tpu-hardware-claim) Anda untuk mengakses hardware TPU mentah secara aman tanpa memerlukan pemasangan volume host yang tidak aman. Terakhir, server API yang kompatibel dengan OpenAI akan diekspos melalui port 8080. Jalankan perintah berikut untuk membuat file:
cat << 'EOF' > gemma-inference.yaml
apiVersion: apps/v1
kind: Deployment
metadata:
  name: vllm-gemma-4
  labels:
    app: gemma-server
spec:
  replicas: 1
  selector:
    matchLabels:
      app: gemma-server
  template:
    metadata:
      labels:
        app: gemma-server
    spec:
      hostIPC: true
      containers:
      - name: vllm-tpu
        image: vllm/vllm-tpu:latest
        securityContext:
          privileged: true
        env:
        - name: HF_TOKEN
          valueFrom:
            secretKeyRef:
              name: hf-token
              key: token
        - name: JAX_PLATFORMS
          value: "tpu,cpu"
        - name: TPU_ACCELERATOR_TYPE
          value: "v6e-4"
        - name: TPU_WORKER_HOSTNAMES
          value: "127.0.0.1"
        - name: TPU_WORKER_ID
          value: "0"
        - name: LIBTPU_INIT_ARGS
          value: "--noenable_tpunetd_client"
        - name: BARE_METAL_MODE
          value: "true"
        - name: BYPASS_VBAR_CONTROL_SERVICE
          value: "1"
        - name: TPU_SKIP_MDS_QUERY
          value: "1"
        - name: TPU_DEFAULT_NETWORK_TYPE
          value: "loopback"
        - name: CHIPS_PER_HOST_BOUNDS
          value: "2,2,1"
        - name: HOST_BOUNDS
          value: "1,1,1"
        - name: ALT
          value: "false,false,false"
        - name: WRAP
          value: "false,false,false"
        command:
        - bash
        - -c
        - |
          export PYTHONUNBUFFERED=1
          sysctl -w net.ipv6.conf.all.disable_ipv6=0
          sysctl -w net.ipv6.conf.default.disable_ipv6=0
          sysctl -w net.ipv6.conf.lo.disable_ipv6=0
          ip link set lo up || true
          
          exec python3 -m vllm.entrypoints.openai.api_server \
            --model google/gemma-4-E4B-it \
            --tensor-parallel-size 4 \
            --trust-remote-code \
            --max-model-len 8192 \
            --max-num-batched-tokens 4096 \
            --host 0.0.0.0 \
            --port 8080
        ports:
        - containerPort: 8080
        resources:
          requests:
            cpu: "170"
            memory: "650Gi"
          limits:
            cpu: "170"
            memory: "650Gi"
          claims:
          - name: tpu-net-claim
          - name: tpu-hardware-claim
        volumeMounts:
        - name: dshm
          mountPath: /dev/shm
      volumes:
      - name: dshm
        emptyDir:
          medium: Memory
      resourceClaims:
      - name: tpu-net-claim
        resourceClaimTemplateName: tpu-net-interfaces
      - name: tpu-hardware-claim
        resourceClaimTemplateName: tpu-device-template
---
apiVersion: v1
kind: Service
metadata:
  name: vllm-gemma-service
spec:
  selector:
    app: gemma-server
  ports:
  - protocol: TCP
    port: 8080
    targetPort: 8080
  type: ClusterIP
EOF
  1. Men-deploy workload Inferensi
kubectl apply -f gemma-inference.yaml
  1. Verifikasi status deployment. Penyiapan ini harus mendownload model dan memuat vLLM. This dapat memakan waktu antara 10 - 25 minutes.
kubectl get pods -l app=gemma-server

kubectl describe pods -l app=gemma-server

Anda juga dapat melihat log dari penampung untuk melihat prosesnya. Tekan CTRL+C untuk keluar dari tampilan log.

kubectl logs -l app=gemma-server -f

Anda akan mengetahui bahwa mesin telah diinisialisasi sepenuhnya saat Anda melihat garis

(APIServer pid=1) INFO: Started server process [1]

(APIServer pid=1) INFO: Waiting for application startup.

(APIServer pid=1) INFO: Application startup complete.

Tekan CTRL+C untuk keluar dari aliran log sebelum melanjutkan.

  1. Verifikasi Lampiran Antarmuka. Periksa antarmuka jaringan yang terikat di dalam container
kubectl exec deployment/vllm-gemma-4 -c vllm-tpu -- ls /sys/class/net
kubectl exec deployment/vllm-gemma-4 -c vllm-tpu -- cat /proc/net/fib_trie | grep -B 1 "32 host"

Yang harus dicari: Anda akan melihat ens9 dan ens10 (atau nama ensX serupa) bersama dengan antarmuka CNI standar (eth0) dan loopback (lo). Antarmuka ini merepresentasikan antarmuka jaringan PCI host GCE fisik yang terikat secara dinamis di dalam pod Anda oleh driver DRANET open source menggunakan konvensi penamaan slot yang dapat diprediksi systemd.

kubectl exec deployment/vllm-gemma-4 -c vllm-tpu -- ls /sys/class/net
ens10
ens9
eth0
Lo

kubectl exec deployment/vllm-gemma-4 -c vllm-tpu -- cat /proc/net/fib_trie | grep -B 1 "32 host"

              |-- 10.10.0.3
                 /32 host LOCAL
--
              |-- 10.20.0.3
                 /32 host LOCAL
--
           |-- 127.0.0.1
              /32 host LOCAL
--
     |-- 192.168.238.67
        /32 host LOCAL
--
              |-- 10.10.0.3
                 /32 host LOCAL
--
              |-- 10.20.0.3
                 /32 host LOCAL
--
           |-- 127.0.0.1
              /32 host LOCAL
--
     |-- 192.168.238.67
        /32 host LOCAL

10. Menguji LLM

Setelah antarmuka Anda divalidasi, luncurkan penampung pengujian ringan di dalam cluster Anda untuk mengirim permintaan inferensi streaming terhadap Gemma 4.

  1. Jalankan perintah berikut pada sesi k8s-control-plane untuk meluncurkan klien interaktif:
kubectl run gemma-chat --rm -i --tty --image=alpine --restart=Never -- sh -c '
  # 1. Silently install curl and jq
  apk add --no-cache curl jq > /dev/null

  echo -e "\n========================================================"
  echo -e "💬 Welcome to the Gemma 4 Real-Time CLI Chat client!"
  echo -e "========================================================"
  echo -e "   Type your prompt below. Type '\''exit'\'' or '\''quit'\'' to end."
  echo -e "========================================================\n"

  while true; do
    # Read user input
    echo -n -e "👤 \033[1;34mYou:\033[0m "
    read -r USER_INPUT
    
    # Handle exit conditions
    if [ "$USER_INPUT" = "exit" ] || [ "$USER_INPUT" = "quit" ] || [ -z "$USER_INPUT" ]; then
      echo -e "\n👋 Goodbye!"
      break
    fi

    echo -n -e "🤖 \033[1;32mGemma:\033[0m "

    # Use jq to safely escape double quotes and special characters in user input
    JSON_PAYLOAD=$(jq -n --arg msg "$USER_INPUT" '\''{
      model: "google/gemma-4-E4B-it",
      messages: [{role: "user", content: $msg}],
      temperature: 0.7,
      stream: true
    }'\'')

    # Stream the tokens in real-time with a typewriter effect
    curl -s -X POST http://vllm-gemma-service:8080/v1/chat/completions \
      -H "Content-Type: application/json" \
      -d "$JSON_PAYLOAD" | while read -r line; do
        # Extract SSE data streams
        if echo "$line" | grep -q "data:"; then
          DATA_CLEAN=$(echo "$line" | sed "s/^data: //" | tr -d "\r")
          if [ "$DATA_CLEAN" != "[DONE]" ] && [ -n "$DATA_CLEAN" ]; then
            # Parse and print only the token content
            TOKEN=$(echo "$DATA_CLEAN" | jq -r ".choices[0].delta.content // empty" 2>/dev/null)
            echo -n "$TOKEN"
          fi
        fi
      done
    echo -e "\n"
  done
'

Interactive chat

7714607072541e90.png

11. Pembersihan

Pertama, hapus semua beban kerja, secret, dan konfigurasi dari cluster Anda.

Jika Anda masih login ke sesi SSH aman k8s-control-plane, jalankan perintah berikut secara langsung. (Jika Anda sudah keluar, masuk kembali melalui SSH terlebih dahulu):

  1. Hubungkan kembali dengan aman ke VM bidang kontrol dari Cloud Shell. Jika Anda sudah terhubung ke VM ini, lewati langkah ini.
gcloud compute ssh k8s-control-plane \
    --zone=$ZONE \
    --tunnel-through-iap
  1. Membersihkan Resource Kubernetes
# 1. Delete the Gemma 4 deployment and service
kubectl delete -f gemma-inference.yaml --ignore-not-found=true

# 2. Delete the Hugging Face access secret
kubectl delete secret hf-token --ignore-not-found=true

# 3. Delete the open-source DRANET specs and drivers
kubectl delete deviceclass dranet --ignore-not-found=true
kubectl delete resourceclaimtemplate tpu-net-interfaces --ignore-not-found=true
kubectl delete -f https://raw.githubusercontent.com/kubernetes-sigs/dranet/refs/heads/main/install.yaml --ignore-not-found=true

# 4. Uninstall the OSS TPU Hardware Driver
helm uninstall dra-driver-google-tpu -n dra-driver-google-tpu --wait || true
  1. Sekarang, ketik exit dan kembali ke direktori Cloud Shell aktif tempat file Terraform Anda disimpan, lalu hancurkan semua node, jaringan VPC, dan aturan firewall.
# 1. Create the teardown script
cat << 'EOF' > teardown.sh
#!/bin/bash

# The specific networks defined in your Terraform vpc.tf
NETWORKS=(
  "oss-k8s-primary-vpc"
  "oss-tpu-vpc-1"
  "oss-tpu-vpc-2"
)

echo "=== Hunting down and deleting ALL firewall rules for OSS networks ==="

for NETWORK in "${NETWORKS[@]}"; do
    echo "Searching for firewall rules attached to network: $NETWORK..."
    
    # Query GCP for any firewall rule tied to this specific network
    STUCK_RULES=$(gcloud compute firewall-rules list \
        --filter="network:($NETWORK)" \
        --format="value(name)" | tr '\n' ' ')
    
    # Check if the string is not empty and contains more than just whitespace
    if [ -n "$STUCK_RULES" ] && [ "$STUCK_RULES" != " " ]; then
        echo "🔥 Found rules holding $NETWORK hostage: $STUCK_RULES"
        echo "Deleting them now..."
        gcloud compute firewall-rules delete $STUCK_RULES --quiet
    else
        echo "✅ No firewall rules found for $NETWORK."
    fi
done

# Fallback: Explicitly delete the named rules from your Terraform file 
# just in case the dynamic filter missed them due to caching delays
echo "=== Running fallback deletion for explicitly named Terraform rules ==="
gcloud compute firewall-rules delete \
    oss-k8s-primary-allow-internal \
    oss-k8s-allow-iap-ssh \
    oss-tpu1-allow-internal \
    oss-tpu2-allow-internal \
    --quiet 2>/dev/null || true

echo "--------------------------------------------------------"
echo "✅ Firewall cleanup complete!"
echo "Your networks are now stripped of firewalls and ready to be deleted."
echo "--------------------------------------------------------"

echo "=== Destroying Infrastructure ==="
cd ~/oss-kube-dra || exit
terraform destroy -auto-approve

echo "--------------------------------------------------------"
echo "✅ Infrastructure successfully destroyed!"
echo "--------------------------------------------------------"
EOF

# 2. Make the script executable and run it
chmod +x teardown.sh
./teardown.sh
  1. Hapus folder terraform oss-kube-dra
cd
rm -r oss-kube-dra

12. Selamat

Anda telah berhasil menyediakan, mem-bootstrap, dan memvalidasi infrastruktur AI Kubernetes berperforma tinggi yang dikelola sendiri langsung di instance VM Google Compute Engine (GCE).

Sekarang Anda memiliki pemahaman mendalam di tingkat sistem tentang cara Kubernetes menggunakan Alokasi Resource Dinamis (DRA) untuk mengatur akselerator TPU mentah, mengikat topologi host multi-NIC berkecepatan tinggi, dan menyajikan model bahasa besar yang canggih.

Langkah berikutnya/Pelajari lebih lanjut

Anda dapat membaca lebih lanjut tentang jaringan GKE

Ikuti lab berikutnya

Lanjutkan quest Anda dengan Google Cloud, dan pelajari lab Google Cloud lainnya: