Autoscale applications based on external events with KEDA

Slides used during the k8s UG on 14 September 2023

Transcript

1. Autoscale Applications based on external events with KEDA Nilesh Gule @nileshgule
2. $whoami { “name” : “Nilesh Gule”, “website” : “https://www.HandsOnArchitect.com", “github” : “https://GitHub.com/NileshGule" “twitter” : “@nileshgule”, “linkedin” : “https://www.linkedin.com/in/nileshgule”, “YouTube” : “https://www.YouTube.com/@nilesh-gule” “likes” : “Technical Evangelism, Cricket”, “co-organizer” : “Azure Singapore UG” }
4. @nileshgule KEDA Kubernets based Event Driven Autoscaling Scaling in Kubernetes KEDA Overview TechTalks demo with RabbitMQ Autoscaling with KEDA KEDA Architecture Agenda
5. Pre-requisites • Self contained application with all its dependencies Docker • Orchestrates containers • Self healing • Service discovery • Scaling Kubernetes • Dockerhub / private container registry • Repository for storing container images Container Registry • Managed Kubernetes service on Azure • Native integration with other Azure services like Azure Active Directory (AAD), Azure Container registry (ACR), Azure Container Instances (ACI), Azure Monitor etc. Azure Kubernetes Service (AKS)
6. Kubernetes scaling options Horizontal Pod Autoscaler Vertical Pod Autoscaler Cluster Autoscaler Manual
7. TechTalks Application Architecture TechTalks Consumer TechTalks Producer TechTalks Frontend (UI) Message Broker Polyglot Programming
8. Dapr overview
9. Techtalks-producer Azure-Singapore-cluster demo-azure-singapore-rg ngacrregistry acrResourceGroup Techtalks-consumer TechTalks with Dapr - AKS
10. @nileshgule Demo KEDA in action
11. KEDA Architecture
12. ScaledObject apiVersion: keda.k8s.io/v1alpha1 kind: ScaledObject metadata: name: rabbitmq-consumer namespace: default labels: deploymentName: rabbitmq-consumer-deployment spec: scaleTargetRef: deploymentName: rabbitmq-consumer pollingInterval: 5 # Optional. Default: 30 seconds cooldownPeriod: 30 # Optional. Default: 300 seconds minReplicaCount: 1 # Optional. Default: 0 maxReplicaCount: 30 # Optional. Default: 100 triggers: - type: rabbitmq metadata: queueName: hello queueLength : ‘5’ authenticationRef: name: trigger-auth-rabbitmq-host
13. KEDA Features
14. Run Containers at scale ❖KEDA supported scale triggers including Azure Event Hub, Apache Kafka, RabbitMQ, MongoDB, MySQL, PostgreSQL etc. ❖Robust autoscaling without managing complex infrastructure ❖Scale to zero and pay only for use by the second https://keda.sh/docs/2.11/scalers/
15. Summary • Modern applications are loosely coupled and highly portable • AKS provides native integrations to Dapr and KEDA • KEDA helps to auto scale on metrics external to Kubernetes • Dapr tries to simplify the Microservices development and deployment • Dapr Components help to extract underlying functionality and provides abstractions • Make app portable to run in serverless as well as managed cloud services
16. References https://www.youtube.com/@nilesh-gule https://dapr.io/ Dapr Publish and Subscribe Kubernetes Event Driven Autoscaling Serverless - Dapr and Azure Container Apps Comparison between ACA and AKS
17. Containerize Apps Resources Cloud Native Ninja : https://github.com/NileshGule/cloud-native-ninja Azure Container Apps : https://github.com/NileshGule/techtalks-azure-container-apps-demo Slides Slideshare: https://www.slideshare.net/nileshgule/ Speaker Deck: https://speakerdeck.com/nileshgule/
18. Nilesh Gule ARCHITECT | MICROSOFT MVP “Code with Passion and Strive for Excellence” nileshgule @nileshgule Nilesh Gule NileshGule www.handsonarchitect.com https://www.youtube.com/@nilesh-gule
19. #GlobalAzure Q&A