Add test cluster autoscaler

helm-charts/README.md

@@ -9,7 +9,8 @@ Helm is a package manager for Kubernetes, simplifying the process of defining, i
 3. [Model Repository S3](#model-repository-s3)
 4. [Install aws-ebs-csi-driver](#install-aws-ebs-csi-driver)
 5. [Install Metric Server](#install-metric-server)
-6. [Install Charts](#install-charts)
+6. [Install Cluster Autoscaler](#install-cluster-autoscaler)
+7. [Install Charts](#install-charts)
    - [Ingress Nginx Controller](#ingress-nginx-controller)
    - [Postgresql](#postgresql)
    - [Elastic Search](#elastic-search)
@@ -20,13 +21,14 @@ Helm is a package manager for Kubernetes, simplifying the process of defining, i
    - [Text Search Application](#text-search-application)
    - [Backend Application](#backend-application)
    - [Frontend Application](#frontend-application)
-7. [Load Test Horizontal Pod Autoscaling](#load-test-horizontal-pod-autoscaling)
+8. [Load Test Autoscaling](#load-test-autoscaling)
    - [Test Backend Horizontal Pod Autoscaling](#test-backend-horizontal-pod-autoscaling)
-8. [Upgrade Charts](#upgrade-charts)
-9. [Uninstall Charts](#uninstall-charts)
-10. [Clean up PVC](#clean-up-pvc)
-11. [Check Resources](#check-resources)
-12. [References](#references)
+   - [Test Cluster Autoscaler ](#test-cluster-autoscaler)
+9. [Upgrade Charts](#upgrade-charts)
+10. [Uninstall Charts](#uninstall-charts)
+11. [Clean up PVC](#clean-up-pvc)
+12. [Check Resources](#check-resources)
+13. [References](#references)
 
 ## Architecture
 
@@ -124,6 +126,14 @@ kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/late
    helm upgrade --install metrics-server metrics-server/metrics-server
    ```
 
+## Install Cluster Autoscaler
+On AWS, Cluster Autoscaler utilizes Amazon EC2 Auto Scaling Groups to manage node groups. Cluster Autoscaler typically runs as a Deployment in your cluster.
+
+Create a Cluster Autoscaler deployment and service account:
+   ```bash
+   kubectl apply -f https://raw.githubusercontent.com/kubernetes/autoscaler/master/cluster-autoscaler/cloudprovider/aws/examples/cluster-autoscaler-autodiscover.yaml
+   ```
+
 ## Install Charts
 Step-by-step instructions to create namespaces and install various Helm charts like Ingress Nginx Controller, Postgresql, Elastic Search, Qdrant, Prometheus, Grafana, and others.
 
@@ -215,7 +225,7 @@ Step-by-step instructions to create namespaces and install various Helm charts l
    helm install frontend-app ./frontend --namespace application --set nodeSelector.nodegroup-type=cpu-nodegroup
    ```
 
-## Load Test Horizontal Pod Autoscaling
+## Load Test Autoscaling
 
 ### Test Backend Horizontal Pod Autoscaling
 
@@ -274,6 +284,15 @@ Step-by-step instructions to create namespaces and install various Helm charts l
       <em>Fig: Locust Test</em>
       </p>
 
+### Test Cluster Autoscaler 
+
+   <p align="center">
+   <img src="./assets/cluster-autoscaler.png" alt="Cluster Autoscaler" />
+   <br>
+   <em>Fig: Cluster Autoscaler</em>
+   </p>
+
+
 ## Upgrade Charts
 Instructions on how to upgrade existing Helm Chart releases.
 
@@ -365,4 +384,6 @@ Deleting PVCs is irreversible and can lead to data loss. Ensure backups are in p
 - [Qdrant Helm Charts](https://github.com/qdrant/qdrant-helm/tree/main/charts/qdrant)
 - [Elastic Cloud on Kubernetes Documentation](https://www.elastic.co/guide/en/cloud-on-k8s/current/k8s-stack-helm-chart.html)
 - [CSI driver for Amazon EBS](https://github.com/kubernetes-sigs/aws-ebs-csi-driver)
-- [Kubernetes Metrics Server](https://github.com/kubernetes-sigs/metrics-server)
+- [Kubernetes Metrics Server](https://github.com/kubernetes-sigs/metrics-server)
+- [Autoscaling in Kubernetes](https://kubernetes.io/blog/2016/07/autoscaling-in-kubernetes/)
+- [Cluster Autoscaler on AWS](https://github.com/kubernetes/autoscaler/blob/master/cluster-autoscaler/cloudprovider/aws/README.md)

helm-charts/backend/templates/deployment.yaml

@@ -57,7 +57,6 @@ spec:
               port: http
             initialDelaySeconds: 5
             periodSeconds: 5
-          # Only limit resources for test hpa
           resources:
             {{- toYaml .Values.resources | nindent 12 }}
       {{- with .Values.nodeSelector }}

helm-charts/backend/values.yaml

@@ -37,8 +37,11 @@ env:
         key: JWT_SECRET
 
 
-# Only limit resources for test hpa
 resources: 
+  # We usually recommend not to specify default resources and to leave this as a conscious
+  # choice for the user. This also increases chances charts run on environments with little
+  # resources, such as Minikube. If you do want to specify resources, uncomment the following
+  # lines, adjust them as necessary, and remove the curly braces after 'resources:'.
   limits:
     cpu: 500m
     memory: 256Mi

helm-charts/cluster-config-eksctl.yaml

@@ -36,26 +36,26 @@ nodeGroups:
     labels:
       nodegroup-type: cpu-nodegroup
 
-  - name: gpu-nodegroup
-    instanceType: g4dn.xlarge
-    instanceName: eks-gpu-node
-    minSize: 1
-    maxSize: 3
-    volumeSize: 60
-    volumeType: gp3
-    desiredCapacity: 1
-    iam:
-      withAddonPolicies:
-        imageBuilder: true
-        autoScaler: true
-        externalDNS: true
-        certManager: true
-        appMesh: true
-        appMeshPreview: true
-        ebs: true
-        fsx: true
-        efs: true
-    labels:
-      nodegroup-type: gpu-nodegroup
+  # - name: gpu-nodegroup
+  #   instanceType: g4dn.xlarge
+  #   instanceName: eks-gpu-node
+  #   minSize: 1
+  #   maxSize: 3
+  #   volumeSize: 60
+  #   volumeType: gp3
+  #   desiredCapacity: 1
+  #   iam:
+  #     withAddonPolicies:
+  #       imageBuilder: true
+  #       autoScaler: true
+  #       externalDNS: true
+  #       certManager: true
+  #       appMesh: true
+  #       appMeshPreview: true
+  #       ebs: true
+  #       fsx: true
+  #       efs: true
+  #   labels:
+  #     nodegroup-type: gpu-nodegroup
 
 availabilityZones: ["us-east-1a", "us-east-1b", "us-east-1c", "us-east-1d"]

helm-charts/frontend/templates/deployment.yaml

@@ -39,6 +39,8 @@ spec:
             httpGet:
               path: /
               port: http
+          resources:
+            {{- toYaml .Values.resources | nindent 12 }}
       {{- with .Values.nodeSelector }}
       nodeSelector:
         {{- toYaml . | nindent 8 }}

helm-charts/frontend/values.yaml

@@ -6,6 +6,14 @@ image:
   pullPolicy: Always
   tag: "latest"
 
+resources: {}
+  # limits:
+  #   cpu: 100m
+  #   memory: 128Mi
+  # requests:
+  #   cpu: 100m
+  #   memory: 128Mi
+
 service:
   type: ClusterIP
   port: 3000

helm-charts/image-search/templates/deployment.yaml

@@ -57,6 +57,8 @@ spec:
           #     port: http
           #   initialDelaySeconds: 5
           #   periodSeconds: 5
+          resources:
+            {{- toYaml .Values.resources | nindent 12 }}
       {{- with .Values.nodeSelector }}
       nodeSelector:
         {{- toYaml . | nindent 8 }}

helm-charts/image-search/values.yaml

@@ -6,6 +6,14 @@ image:
   pullPolicy: Always
   tag: "latest"
 
+resources: {}
+  # limits:
+  #   cpu: 100m
+  #   memory: 128Mi
+  # requests:
+  #   cpu: 100m
+  #   memory: 128Mi
+
 service:
   type: ClusterIP
   port: 7000

helm-charts/text-search/templates/deployment.yaml

@@ -57,6 +57,8 @@ spec:
           #     port: http
           #   initialDelaySeconds: 5
           #   periodSeconds: 5
+          resources:
+            {{- toYaml .Values.resources | nindent 12 }}
       {{- with .Values.nodeSelector }}
       nodeSelector:
         {{- toYaml . | nindent 8 }}

helm-charts/text-search/values.yaml

@@ -6,6 +6,14 @@ image:
   pullPolicy: Always
   tag: "latest"
 
+resources: {}
+  # limits:
+  #   cpu: 100m
+  #   memory: 128Mi
+  # requests:
+  #   cpu: 100m
+  #   memory: 128Mi
+
 service:
   type: ClusterIP
   port: 6000