This page shows you how to request hardware accelerators (GPUs) in your Google Kubernetes Engine (GKE) Autopilot workloads.
GPUs in regular Autopilot Pods are eligible for Committed Use Discounts (CUDs).
Before you begin
Before you start, make sure you have performed the following tasks:
- Enable the Google Kubernetes Engine API. Enable Google Kubernetes Engine API
- If you want to use the Google Cloud CLI for this task, install and then initialize the gcloud CLI.
- Ensure that you have a GKE Autopilot cluster running GKE version 1.24.2-gke.1800 or later.
- You can't use time-sharing GPUs and multi-instance GPUs with Autopilot.
- GPU availability depends on the Google Cloud region of your Autopilot cluster, and your GPU quota.
- If you explicitly request a specific existing GPU node for your Pod, the Pod must consume all the GPU resources on the node. For example, if the existing node has 8 GPUs and your Pod's containers request a total of 4 GPUs, Autopilot rejects the Pod.
Request GPUs in your containers
To request GPU resources for your containers, add the following fields to your Pod specification:
apiVersion: v1 kind: Pod metadata: name: my-gpu-pod spec: nodeSelector: cloud.google.com/gke-accelerator: GPU_TYPE containers: - name: my-gpu-container image: nvidia/cuda:11.0.3-runtime-ubuntu20.04 command: ["/bin/bash", "-c", "--"] args: ["while true; do sleep 600; done;"] resources: limits: nvidia.com/gpu: GPU_QUANTITY
Replace the following:
GPU_TYPE: the type of GPU hardware. Allowed values are the following:
GPU_QUANTITY: the number of GPUs to allocate to the container. Must be a supported GPU quantity for the GPU type you selected.
You must specify both the GPU type and the GPU quantity in your Pod specification. If you omit either of these values, Autopilot rejects your Pod.
CPU and memory requests for Autopilot GPU Pods
When defining your GPU Pods, you should also request CPU and memory resources so that your containers perform as expected. Autopilot enforces specific CPU and memory minimums, maximums, and defaults based on the GPU type and quantity. For details, refer to Resource requests in Autopilot.
Your Pod specification should look similar to the following example, which requests four T4 GPUs:
apiVersion: v1 kind: Pod metadata: name: t4-pod spec: nodeSelector: cloud.google.com/gke-accelerator: "nvidia-tesla-t4" containers: - name: t4-container-1 image: nvidia/cuda:11.0.3-runtime-ubuntu20.04 command: ["/bin/bash", "-c", "--"] args: ["while true; do sleep 600; done;"] resources: limits: nvidia.com/gpu: 3 requests: cpu: "54" memory: "54Gi" - name: t4-container-2 image: nvidia/cuda:11.0.3-runtime-ubuntu20.04 command: ["/bin/bash", "-c", "--"] args: ["while true; do sleep 600; done;"] resources: limits: nvidia.com/gpu: 1 requests: cpu: "18" memory: "18Gi"
Verify GPU allocation
To check that a deployed GPU workload has the requested GPUs, run the following command:
kubectl describe node NODE_NAME
NODE_NAME with the name of the node on which the
Pod was scheduled.
The output is similar to the following:
apiVersion: v1 kind: Node metadata: ... labels: ... cloud.google.com/gke-accelerator: nvidia-tesla-t4 cloud.google.com/gke-accelerator-count: "1" cloud.google.com/machine-family: custom-48 ... ...
How GPU allocation works in Autopilot
After you request a GPU type and a quantity for the containers in a Pod and deploy the Pod, the following happens:
- If no allocatable GPU node exists, Autopilot provisions a new GPU node to schedule the Pod. Autopilot automatically installs NVIDIA's drivers to facilitate the hardware.
- Autopilot adds node taints to the GPU node and the corresponding tolerations to the Pod. This prevents GKE from scheduling other Pods on the GPU node.
Autopilot places exactly one GPU Pod on each GPU node, as well as any GKE-managed workloads that run on all nodes, and any DaemonSets that you configure to tolerate all node taints.
Run DaemonSets on every node
You might want to run DaemonSets on every node, even nodes with applied taints. For example, some logging and monitoring agents must run on every node in the cluster. You can configure those DaemonSets to ignore node taints so that GKE places those workloads on every node.
To run DaemonSets on every node in your cluster, including your GPU nodes, add the following toleration to your specification:
apiVersion: apps/v1 kind: DaemonSet metadata: name: logging-agent spec: tolerations: - key: "" operator: "Exists" effect: "" containers: - name: logging-agent-v1 image: IMAGE_PATH
To run DaemonSets on specific GPU nodes in your cluster, add the following to your specification:
apiVersion: apps/v1 kind: DaemonSet metadata: name: logging-agent spec: nodeSelector: cloud.google.com/gke-accelerator: "GPU_TYPE" tolerations: - key: "" operator: "Exists" effect: "" containers: - name: logging-agent-v1 image: IMAGE_PATH
GPU_TYPE with the type of GPU in your target
nodes. Can be either
GPU use cases in Autopilot
You can allocate GPUs to containers in Autopilot Pods to facilitate workloads such as the following:
- Machine learning (ML) inference
- ML training
Supported GPU quantities
When you request GPUs in your Pod specification, you must use the following quantities based on the GPU type:
||1, 2, 4|
||1, 2, 4, 8, 16|
If you request a GPU quantity that isn't supported for that type, Autopilot rejects your Pod.
Monitor GPU nodes
- Duty Cycle (
container/accelerator/duty_cycle): Percentage of time over the past sample period (10 seconds) during which the accelerator was actively processing. Between 1 and 100.
- Memory Usage (
container/accelerator/memory_used): Amount of accelerator memory allocated in bytes.
- Memory Capacity (
container/accelerator/memory_total): Total accelerator memory in bytes.
For more information about monitoring your clusters and their resources, refer to Monitoring.
View usage metrics
You view your workload GPU usage metrics from the Workloads dashboard in theGoogle Cloud console.
To view your workload GPU usage, perform the following steps:
Go to the Workloads page in the Google Cloud console.Go to Workloads
- Select a workload.
The Workloads dashboard displays charts for GPU memory usage and capacity, and GPU duty cycle.
- Learn more about GPU support in GKE.
- Read about how Autopilot compute classes are optimized for specialized use cases.