Assign Pod-level CPU and memory resources
Kubernetes v1.32 [alpha] (enabled by default: false)
This page shows how to specify CPU and memory resources for a Pod at pod-level in addition to container-level resource specifications. A Kubernetes node allocates resources to a pod based on the pod's resource requests. These requests can be defined at the pod level or individually for containers within the pod. When both are present, the pod-level requests take precedence.
Similarly, a pod's resource usage is restricted by limits, which can also be set at the pod-level or individually for containers within the pod. Again, pod-level limits are prioritized when both are present. This allows for flexible resource management, enabling you to control resource allocation at both the pod and container levels.
In order to specify the resources at pod-level, it is required to enable
PodLevelResources feature gate.
For Pod Level Resources:
- Priority: When both pod-level and container-level resources are specified, pod-level resources take precedence.
- QoS: Pod-level resources take precedence in influencing the QoS class of the pod.
- OOM Score: The OOM score adjustment calculation considers both pod-level and container-level resources.
- Compatibility: Pod-level resources are designed to be compatible with existing features.
Before you begin
You need to have a Kubernetes cluster, and the kubectl command-line tool must be configured to communicate with your cluster. It is recommended to run this tutorial on a cluster with at least two nodes that are not acting as control plane hosts. If you do not already have a cluster, you can create one by using minikube or you can use one of these Kubernetes playgrounds:
Your Kubernetes server must be at or later than version 1.32. To check the version, enterkubectl version.
The PodLevelResources feature
gate must be enabled
for your control plane and for all nodes in your cluster.
Create a namespace
Create a namespace so that the resources you create in this exercise are isolated from the rest of your cluster.
kubectl create namespace pod-resources-example
Create a pod with memory requests and limits at pod-level
To specify memory requests for a Pod at pod-level, include the resources.requests.memory
field in the Pod spec manifest. To specify a memory limit, include resources.limits.memory.
In this exercise, you create a Pod that has one Container. The Pod has a memory request of 100 MiB and a memory limit of 200 MiB. Here's the configuration file for the Pod:
apiVersion: v1
kind: Pod
metadata:
name: memory-demo
namespace: pod-resources-example
spec:
resources:
requests:
memory: "100Mi"
limits:
memory: "200Mi"
containers:
- name: memory-demo-ctr
image: nginx
command: ["stress"]
args: ["--vm", "1", "--vm-bytes", "150M", "--vm-hang", "1"]
The args section in the manifest provides arguments for the container when it starts.
The "--vm-bytes", "150M" arguments tell the Container to attempt to allocate 150 MiB of memory.
Create the Pod:
kubectl apply -f https://k8s.io/examples/pods/resource/pod-level-memory-request-limit.yaml --namespace=pod-resources-example
Verify that the Pod is running:
kubectl get pod memory-demo --namespace=pod-resources-example
View detailed information about the Pod:
kubectl get pod memory-demo --output=yaml --namespace=pod-resources-example
The output shows that the Pod has a memory request of 100 MiB and a memory limit of 200 MiB.
...
spec:
containers:
...
resources:
requests:
memory: 100Mi
limits:
memory: 200Mi
...
Run kubectl top to fetch the metrics for the pod:
kubectl top pod memory-demo --namespace=pod-resources-example
The output shows that the Pod is using about 162,900,000 bytes of memory, which is about 150 MiB. This is greater than the Pod's 100 MiB request, but within the Pod's 200 MiB limit.
NAME CPU(cores) MEMORY(bytes)
memory-demo <something> 162856960
Create a pod with CPU requests and limits at pod-level
To specify a CPU request for a Pod, include the resources.requests.cpu field
in the Pod spec manifest. To specify a CPU limit, include resources.limits.cpu.
In this exercise, you create a Pod that has one container. The Pod has a request of 0.5 CPU and a limit of 1 CPU. Here is the configuration file for the Pod:
apiVersion: v1
kind: Pod
metadata:
name: cpu-demo
namespace: pod-resources-example
spec:
resources:
limits:
cpu: "1"
requests:
cpu: "0.5"
containers:
- name: cpu-demo-ctr
image: vish/stress
args:
- -cpus
- "2"
The args section of the configuration file provides arguments for the container when it starts.
The -cpus "2" argument tells the Container to attempt to use 2 CPUs.
Create the Pod:
kubectl apply -f https://k8s.io/examples/pods/resource/pod-level-cpu-request-limit.yaml --namespace=pod-resources-example
Verify that the Pod is running:
kubectl get pod cpu-demo --namespace=pod-resources-example
View detailed information about the Pod:
kubectl get pod cpu-demo --output=yaml --namespace=pod-resources-example
The output shows that the Pod has a CPU request of 500 milliCPU and a CPU limit of 1 CPU.
spec:
containers:
...
resources:
limits:
cpu: "1"
requests:
cpu: 500m
Use kubectl top to fetch the metrics for the Pod:
kubectl top pod cpu-demo --namespace=pod-resources-example
This example output shows that the Pod is using 974 milliCPU, which is slightly less than the limit of 1 CPU specified in the Pod configuration.
NAME CPU(cores) MEMORY(bytes)
cpu-demo 974m <something>
Recall that by setting -cpu "2", you configured the Container to attempt to use 2
CPUs, but the Container is only being allowed to use about 1 CPU. The container's
CPU use is being throttled, because the container is attempting to use more CPU
resources than the Pod CPU limit.
Create a pod with resource requests and limits at both pod-level and container-level
To assign CPU and memory resources to a Pod, you can specify them at both the pod
level and the container level. Include the resources field in the Pod spec to
define resources for the entire Pod. Additionally, include the resources field
within container's specification in the Pod's manifest to set container-specific
resource requirements.
In this exercise, you'll create a Pod with two containers to explore the interaction of pod-level and container-level resource specifications. The Pod itself will have defined CPU requests and limits, while only one of the containers will have its own explicit resource requests and limits. The other container will inherit the resource constraints from the pod-level settings. Here's the configuration file for the Pod:
apiVersion: v1
kind: Pod
metadata:
name: pod-resources-demo
namespace: pod-resources-example
spec:
resources:
limits:
cpu: "1"
memory: "200Mi"
requests:
cpu: "1"
memory: "100Mi"
containers:
- name: pod-resources-demo-ctr-1
image: nginx
resources:
limits:
cpu: "0.5"
memory: "100Mi"
requests:
cpu: "0.5"
memory: "50Mi"
- name: pod-resources-demo-ctr-2
image: fedora
command:
- sleep
- inf
Create the Pod:
kubectl apply -f https://k8s.io/examples/pods/resource/pod-level-resources.yaml --namespace=pod-resources-example
Verify that the Pod Container is running:
kubectl get pod-resources-demo --namespace=pod-resources-example
View detailed information about the Pod:
kubectl get pod memory-demo --output=yaml --namespace=pod-resources-example
The output shows that one container in the Pod has a memory request of 50 MiB and a CPU request of 0.5 cores, with a memory limit of 100 MiB and a CPU limit of 0.5 cores. The Pod itself has a memory request of 100 MiB and a CPU request of 1 core, and a memory limit of 200 MiB and a CPU limit of 1 core.
...
containers:
name: pod-resources-demo-ctr-1
resources:
requests:
cpu: 500m
memory: 50Mi
limits:
cpu: 500m
memory: 100Mi
...
name: pod-resources-demo-ctr-2
resources: {}
resources:
limits:
cpu: 1
memory: 200Mi
requests:
cpu: 1
memory: 100Mi
...
Since pod-level requests and limits are specified, the request guarantees for both containers in the pod will be equal 1 core or CPU and 100Mi of memory. Additionally, both containers together won't be able to use more resources than specified in the pod-level limits, ensuring they cannot exceed a combined total of 200 MiB of memory and 1 core of CPU.
Clean up
Delete your namespace:
kubectl delete namespace pod-resources-example