quarkus-vs-springboot

A demo project that demonstrates the differences and similarities between the frameworks

Features:

• Controller
  • REST interface to read all orders
  • REST interface to post a new order
  • New orders are send using Kafka to the order service
  • Existing orders are gotten trough HTTP from the order service
• Service
  • Postgres database
  • Kafka consumer for new orders
  • REST interface for all orders (use paging)

Original creation

The Quarkus projects are created using the Quarkus CLI
The Spring Boot projects are created using the Spring Initializr

Both projects with Gradle and Kotlin options on

needed tools

• Java
• Docker
• Kubectl
• Minikube
• JQ

setup minikube

There is an actions.sh script in /scripts. Use this to perform the following steps:

• ./actions.sh create-mount Make mounts for Minikube to store persistent data
• ./actions.sh start-minikube Start Minikube with the correct settings
• ./actions.sh build-kafka Build a Kafka that works without Zookeeper
• ./actions.sh start-kafka Start Kafka in Minikube
• ./actions.sh start-postgres Start Postgres in Minikube
• ./actions.sh start-ingress Start Ingress in Minikube (if you run this on Apple Silicon then it will create a tunnel for Ingress and not come back with a prompt. (You also probably will be asked to enter the sudo password))
• ./actions.sh refrest-services Build and deploy all services to Minikube

pod scaling demo

There is an actions.sh script in /scripts. Use this to perform the following steps:

• ./actions.sh

First for spring-boot:

• ./actions.sh build-image spring-boot order-controller
• ./actions.sh start-pods spring-boot order-controller
  This will barely be able to start the pods in time and by doing so use a lot of resources.
• ./actions.sh replicas spring-boot order-controller 30
  This will take forever with a lot of CrashLoopBackOff. Better do this in small increments of 4 replicas a time.
• ./actions.sh drain
  This will drain one of the nodes and move all pods to another node at once. See the load increase and wait forever for this to have all pods running again.
• ./actions.sh uncordon Make the node available again
• ./actions.sh replicas spring-boot order-controller 1 Scale back to one

Second for quarkus:

• ./action.sh build-image quarkus order-controller
• ./action.sh start-pods quarkus order-controller
  This will quickly start the pods and use little resources.
• ./action.sh replicas quarkus order-controller 30
  There might be some restarts but it reaches the 30 pods quite quickly.
• ./action.sh drain
  This will drain one of the nodes and move all pods to another node at once. See how quickly this finishes and moves all the drained pods to the other node.
• ./action.sh uncordon Make the node available again
• ./action.sh replicas quarkus order-controller 1 Scale back to one