Bits of .NET

🎯 Put Pods Where They Belong

Some nodes have GPUs. Some are in different zones. Node affinity schedules pods on specific nodes. Control placement precisely.

📝 Node Affinity

apiVersion: v1
kind: Pod
metadata:
  name: gpu-pod
spec:
  affinity:
    nodeAffinity:
      requiredDuringSchedulingIgnoredDuringExecution:
        nodeSelectorTerms:
        - matchExpressions:
          - key: gpu
            operator: In
            values:
            - "nvidia"
            - "amd"
  containers:
  - name: app
    image: myapp:latest

# Also supports:
# - requiredDuringScheduling: Hard requirement
# - preferredDuringScheduling: Soft preference
# - NotIn, Exists, DoesNotExist operators

🎯 Use Cases

# GPU workloads
affinity:
  nodeAffinity:
    requiredDuringScheduling:
      nodeSelectorTerms:
      - matchExpressions:
        - key: gpu
          operator: In
          values: ["nvidia"]

# Zone affinity (high availability)
affinity:
  nodeAffinity:
    preferredDuringScheduling:
    - preference:
        matchExpressions:
        - key: topology.kubernetes.io/zone
          operator: In
          values: ["us-east-1a"]
      weight: 100

# Node with SSD
affinity:
  nodeAffinity:
    requiredDuringScheduling:
      nodeSelectorTerms:
      - matchExpressions:
        - key: storage
          operator: In
          values: ["ssd"]

💡 Node Selector vs Node Affinity

• Node Selector: Simple (key: value)
• Node Affinity: Complex (operators, multiple terms)
• Both support required and preferred
• Use node selector for simple cases
• Use node affinity for complex conditions