Kubernetes Resource Requests, Limits, and Quotas

Kubernetes allows us to manage resources efficiently by defining resource requests and limits for containers. Additionally, **resource quotas** help enforce usage policies at the namespace level, ensuring resources are shared fairly across teams or projects. In this guide, we’ll explore resource requests, limits, and quotas, along with examples of how to configure them in a Kubernetes cluster.

What are Resource Requests?

Resource requests define the minimum amount of CPU and memory a container needs to operate. Kubernetes uses these values when deciding where to schedule a pod.

What are Resource Limits?

Resource limits define the maximum amount of CPU and memory a container can use. If a container tries to exceed these limits, Kubernetes will either throttle the container (for CPU) or terminate it (for memory).

Why Use Resource Requests and Limits?

• Efficient Scheduling: Kubernetes schedules pods based on the requested resources to avoid overloading nodes.
• Prevent Resource Contention: Limits ensure no single container monopolizes resources, leaving enough for other workloads.
• Quality of Service (QoS): Kubernetes uses QoS classes to prioritize pods based on their resource requests and limits.

Defining Resource Requests and Limits

You can define resource requests and limits in a pod's specification, as shown in this example:

apiVersion: v1
kind: Pod
metadata:
  name: myapp-pod
spec:
  containers:
    - name: myapp-container
      image: myapp:latest
      resources:
        requests:
          memory: "64Mi"
          cpu: "250m"
        limits:
          memory: "128Mi"
          cpu: "500m"

What is a Resource Quota?

A **resource quota** is a mechanism that sets hard limits on the amount of resources (like CPU, memory, or storage) a namespace can consume. Resource quotas are useful for limiting resource usage in environments shared by multiple teams or projects. By enforcing quotas, administrators can ensure fair resource distribution across the cluster.

Defining a Resource Quota

You can define a resource quota using a YAML file like the following example:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: dev-quota
  namespace: dev
spec:
  hard:
    pods: "10"
    requests.cpu: "4"
    requests.memory: 8Gi
    limits.cpu: "8"
    limits.memory: 16Gi

In this example, the dev namespace is limited to:

• 10 pods
• 4 CPU cores requested
• 8Gi of memory requested
• 8 CPU cores as the maximum limit
• 16Gi of memory as the maximum limit

Applying a Resource Quota

Once you define the resource quota in the YAML file, apply it to the cluster using the following command:

Example output:

resourcequota/dev-quota created

Viewing Resource Quota Usage

To check how much of the quota has been consumed in a namespace, use the following command:

Example output:

apiVersion: v1
kind: ResourceQuota
metadata:
  name: dev-quota
  namespace: dev
spec:
  hard:
    limits.cpu: "8"
    limits.memory: 16Gi
    pods: "10"
    requests.cpu: "4"
    requests.memory: 8Gi
status:
  hard:
    limits.cpu: "8"
    limits.memory: 16Gi
    pods: "10"
    requests.cpu: "4"
    requests.memory: 8Gi
  used:
    limits.cpu: "2"
    limits.memory: 4Gi
    pods: "5"
    requests.cpu: "1"
    requests.memory: 2Gi

In this output, the hard section shows the quota's limits, and the used section shows the current resource usage. In this case, 5 out of 10 pods are in use, with 2Gi of memory and 1 CPU core requested.

Deleting a Resource Quota

To delete a resource quota from a namespace, use the following command:

Example output:

resourcequota "dev-quota" deleted

Monitoring Resource Usage with kubectl top

You can monitor resource usage in real-time with the following command:

Example output:

NAME              CPU(cores)   MEMORY(bytes)
my-app-1          200m          100Mi
my-app-2          300m          150Mi

This command helps you understand how much CPU and memory each pod is consuming, and you can compare it with the resource requests and limits defined earlier.

Conclusion

Resource requests, limits, and quotas are essential tools for efficient resource management in Kubernetes. By setting proper resource requests and limits at the container level and enforcing quotas at the namespace level, you can ensure fair and efficient use of your cluster's resources.