Horizontal Pod Autoscaler
Horizontal, Pod, Autoscaler, (HPA), Deployment/ReplicaSet, ke, replicas, count, ko, AUTOMATICALLY, ADJUST, karta hai, METRICS, (usually, CPU, ya, memory, UTILIZATION) ke, BASIS, PAR — TRAFFIC, BADHNE, PAR, MORE, pods, GHATNE, PAR, KAM, pods.
HPA, EK, TARGET, METRIC, (jaise, "50%, CPU, utilization") DEFINE, karta hai — agar, ACTUAL, utilization, TARGET, se, ZYAADA, hai, TO, replicas, BADHTE, HAIN, KAM, hai, TO, GHATTE, HAIN (min/max, BOUNDS, ke, ANDAR).
kubectl autoscale deployment nginx --cpu-percent=50 --min=2 --max=10
apiVersion: autoscaling/v2
kind: HorizontalPodAutoscaler
metadata:
name: nginx-hpa
spec:
scaleTargetRef:
apiVersion: apps/v1
kind: Deployment
name: nginx
minReplicas: 2
maxReplicas: 10
metrics:
- type: Resource
resource:
name: cpu
target:
type: Utilization
averageUtilization: 50- HPA = replicas count ko automatically adjust karta hai, metrics ke basis par
- Target metric (jaise 50% CPU) define karo, min/max bounds ke saath
- metrics-server install hona zaroori hai HPA ke kaam karne ke liye
HPA, sirf, CPU/memory, tak, LIMITED, NAHI, hai — Prometheus, Adapter, jaisi, tools, ke, saath, CUSTOM, metrics, (jaise, "requests-per-second" ya, "queue, length") PAR, bhi, autoscale, kiya, ja, sakta hai.
metrics:
- type: Pods
pods:
metric:
name: requests_per_second
target:
type: AverageValue
averageValue: 100HPA, REPLICAS, count, badhata, hai (horizontal, scaling), JABKI, Vertical, Pod, Autoscaler, (VPA), EXISTING, pods, ke, resource, requests/limits, ko, AUTOMATICALLY, ADJUST, karta hai (vertical, scaling) — DONO, KABHI-KABHI, SAATH, mein, bhi, use, kiye, jaate, HAIN.