/sieve-challenge

Primary LanguagePython

How to run the project

There are 4 components that you need to run on kuberentes

  1. Redis
  2. Celery runner
  3. Core API
  4. ML Model

Redis setup:

  • Go into the redis directory and run the following commands
  1. kubectl create -f sc.yaml
  2. kubectl create -f pv.yaml
  3. kubectl create -f redis-configmap.yaml
  4. kubectl create -f redis-statefulset.yaml
  5. kubectl create -f redis-service.yaml

Installing Metrics Server(Needed for horizontal pod autoscaler):

  1. kubectl apply -f https://github.com/kubernetes-sigs/metrics-server/releases/latest/download/components.yaml
  2. kubectl edit deploy -n kube-system metrics-server
  3. Set the arguments to the following:
   args:
     - --kubelet-insecure-tls
     - --kubelet-preferred-address-types=InternalIP,ExternalIP,Hostname

Celery setup:

  • Go into the celery-task-queue/celery_runner directory and run the following commands
  1. kubectl create -f deploy.yaml
  2. kubectl create -f hpa.yaml

Core API setup:

  • Go into the celery-task-queue/core_api directory and run the following commands
  1. kubectl create -f deploy.yaml

ML Model setup:

  • Go into the celery-task-queue/ml_models directory and run the following commands
  1. kubectl create -f deploy.yaml
  2. kubectl create -f hpa.yaml

Port-forward the api service:

  1. kubectl port-forward svc/core-api-layer-service 8002 -n sieve