diff --git a/helm-charts/README.md b/helm-charts/README.md
index d3c62fb..8478a76 100644
--- a/helm-charts/README.md
+++ b/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
Fig: Locust Test
+### Test Cluster Autoscaler
+
+
+ 
+
+ Fig: Cluster Autoscaler
+
+
+
## 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)
\ No newline at end of file
+- [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)
\ No newline at end of file
diff --git a/helm-charts/assets/cluster-autoscaler.png b/helm-charts/assets/cluster-autoscaler.png
new file mode 100644
index 0000000..d8c4f26
Binary files /dev/null and b/helm-charts/assets/cluster-autoscaler.png differ
diff --git a/helm-charts/backend/templates/deployment.yaml b/helm-charts/backend/templates/deployment.yaml
index 9b08a0e..45477ac 100644
--- a/helm-charts/backend/templates/deployment.yaml
+++ b/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 }}
diff --git a/helm-charts/backend/values.yaml b/helm-charts/backend/values.yaml
index 96dcbc6..a6c3958 100644
--- a/helm-charts/backend/values.yaml
+++ b/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
diff --git a/helm-charts/cluster-config-eksctl.yaml b/helm-charts/cluster-config-eksctl.yaml
index e595b1f..04b1214 100644
--- a/helm-charts/cluster-config-eksctl.yaml
+++ b/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"]
diff --git a/helm-charts/frontend/templates/deployment.yaml b/helm-charts/frontend/templates/deployment.yaml
index 46c2ed4..5466e86 100644
--- a/helm-charts/frontend/templates/deployment.yaml
+++ b/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 }}
diff --git a/helm-charts/frontend/values.yaml b/helm-charts/frontend/values.yaml
index 3fb47e0..02b782b 100644
--- a/helm-charts/frontend/values.yaml
+++ b/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
diff --git a/helm-charts/image-search/templates/deployment.yaml b/helm-charts/image-search/templates/deployment.yaml
index 5ead782..4c72a74 100644
--- a/helm-charts/image-search/templates/deployment.yaml
+++ b/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 }}
diff --git a/helm-charts/image-search/values.yaml b/helm-charts/image-search/values.yaml
index 768accf..5c47cc2 100644
--- a/helm-charts/image-search/values.yaml
+++ b/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
diff --git a/helm-charts/text-search/templates/deployment.yaml b/helm-charts/text-search/templates/deployment.yaml
index ea364f4..932448a 100644
--- a/helm-charts/text-search/templates/deployment.yaml
+++ b/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 }}
diff --git a/helm-charts/text-search/values.yaml b/helm-charts/text-search/values.yaml
index cf46922..45f9694 100644
--- a/helm-charts/text-search/values.yaml
+++ b/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