Cloud-native Patterns (July 4th, 2019)
2 likes199 views
The document outlines various cloud-native patterns and best practices for building scalable and resilient applications, focusing on principles like stateless architectures, circuit breakers, and API design. It covers the application of these patterns through demos and discusses tools for service discovery, chaos engineering, and observability. Additionally, it emphasizes the importance of continuous improvement and productivity in software development on a unified platform.
Cloud-native Patterns (July 4th, 2019)
- 1. © Copyright 2019 Pivotal Software, Inc. All rights Reserved. Cloud-native Patterns Didier Burkhalter Alexandre Roman Platform Architect Platform Architect @didierDIB @Alexandre_Roman July 4th, 2019
- 2. Welcome to this session! Didier Burkhalter Platform Architect, Pivotal @didierDIB Alexandre Roman Platform Architect, Pivotal @Alexandre_Roman
- 3. Cover w/ Image Agenda ■ Cloud-native patterns overview ■ Apply these patterns to your apps ■ Demo ■ Q+A during the lunch ;)
- 5. A map for Cloud-native patterns Stateless DDD / Bounded context / fine grain 12 factors Circuit breaker Client LB YP / Service Registry asynchronous API first API gateway event driven disposable (12X) chaos engineering CI/CD Small iterations Continuous refactoring instant scale up / scale down site reliability engineering Continuous update / patch Vault External config’s Contract Data services abstractions anti fragile / resiliency product centric observability SLO / SLI Design Security Behavior Team Distributed Architectures Value via communication
- 6. A map for Cloud-native patterns 12 factors Circuit breaker Client LB YP / Service Registry asynchronous API gateway event driven disposable (12X) chaos engineering CI/CD Small iterations Continuous refactoring instant scale up / scale down site reliability engineering Continuous update / patch Vault External config’s Contract Data services abstractions anti fragile / resiliency product centric observability SLO / SLI Design Security Behavior Team Distributed Architectures Value via communication Stateless DDD / Bounded context / fine grain
- 7. How to apply these patterns to your apps?
- 8. Demo app architecture API Gateway Spring Cloud Gateway Frontend UI Spring Boot + Vue.js Users API Spring Cloud Connections API Spring Cloud Feeds API Spring Cloud Search API Spring Cloud Posts API Spring Cloud Service Discovery Netflix Eureka httpshttps
- 9. Cloud-native patterns in action yatc.cfapps.io
- 10. API first API as a product ➔ Design fine-grained APIs ◆ Balance between chatty and coarse-grained APIs ➔ Apply domain driven design, embrace bounded context principles ➔ Distributed architecture: a way to scale out your application workloads ➔ Composability: build your app from component parts ➔ Hide your microservices implementation using Spring Cloud Gateway ◆ Keep your public API stable, handle breaking changes with care
- 11. API gateway and Service discovery in action API Gateway Spring Cloud Gateway Frontend UI Spring Boot + Vue.js Users API Spring Cloud Connections API Spring Cloud Feeds API Spring Cloud Search API Spring Cloud Posts API Spring Cloud Service Discovery Netflix Eureka httpshttps
- 12. Service discovery Know where you are ➔ Leverage Spring Cloud to locate your services ◆ Pluggable DiscoveryClient extensions: Consul, Kubernetes, Netflix Eureka ➔ Think dynamic, not static: you may scale out your app (more instances), or scale to zero ➔ Build with resilience in mind: apply circuit breaker pattern ◆ You don’t know when a service is down ➔ Spring Cloud provides a high level platform abstraction ◆ WebClient / RestTemplate nicely fit with Netflix OSS, Resilience4j, Kubernetes
- 13. Scaling out your app API Gateway Spring Cloud Gateway Frontend UI Spring Boot + Vue.js Users API Spring Cloud Connections API Spring Cloud Feeds API Spring Cloud Search API Spring Cloud Posts API Spring Cloud Service Discovery Netflix Eureka httpshttps
- 14. Me doing chaos monkey in production
- 15. Chaos engineering “Organize” the chaos, Murphy’s law is coming! Or … Failure is normal ! ➔ Disposability: get help from both your design and platform features ◆ Kill your app and wait for “scream” ◆ Let the platform recover your app and analyse what happens ◆ Scale out your app, kill a container and analyze what happens ➔ Test often, test early, break often, break early (tested last week …) ◆ Functional tests are not enough: the datacenter will burn ◆ More time to enhance your design ➔ Never trust the network ◆ “It was DNS” or the Load balancer or …
- 16. Observability & monitoring Improved measures optimize mission effectiveness: give sense, collect sense, shared sense ➔ Leverage out-of-the-box non-functional instruments ◆ Just add Spring Boot Actuator and Micrometer Prometheus Exporter ➔ Use tailored tools to scrape metrics ◆ Prometheus, Datadog, Dynatrace, Application Insights, Wavefront, etc. ➔ Create ad-hoc dashboards depending on who’s viewing it ◆ Build your own dashboard using Grafana
- 17. Exposing app metrics API gateway Spring Cloud Gateway Frontend UI Spring Boot + Vue.js httpshttps Prometheus Database Grafana Dashboard
- 18. Demo
- 19. Monitoring apps with Prometheus/Grafana
- 20. Monitoring apps with Prometheus/Grafana Monitoring gateway with Prometheus/Grafana
- 21. Spring Cloud Sleuth & .Net Core - OpenTracing combined
- 22. Cloud-native patterns seen as a “mille-feuille” Methodology Learn the necessary practices to build modern software Tools Build with products focused on developer productivity Platform Run every app, on every cloud, on a unified platform Cloud-native software Pivotal Labs and Services Pivotal Cloud Foundry Best-in-class products Your teams Pivotal Culture Continually improve and deliver customer success
- 23. Resources
- 24. Resources ➔ The 12 Factor App manifesto: 12factor.net ➔ Google Site Reliability Engineering: landing.google.com/sre/books ➔ Error budget: www.youtube.com/watch?v=lXPmh9Ap114 ➔ How Improved Measures Optimize Mission Effectiveness white-paper ➔ Get the app source code: github.com/alexandreroman/yatc ➔ Chaos Monkey for Spring Boot codecentric.github.io/chaos-monkey-spring-boot
- 25. Tirage au sort - Des Google Nest Hub Charbon à gagner Remplissez le formulaire d’évaluation avant 15h30
- 26. Transforming how the world builds software © Copyright 2019 Pivotal Software, Inc. All rights Reserved.