Developing Microservices using Spring - Beginner's Guide

Developing
Microservices
using Spring
-
Beginners Guide
* Disclaimer: There are various topics covered in this session, each topic is worth a 2 day workshop session in itself, so this will cover only the very basics and try
to provide a birds eye view of the various concepts.

Agenda
● Microservices
● What ?
● Why ?
● Challenges of a distributed system
● Building Microservices using Spring
● Spring Boot
● Spring Cloud & Netflix OSS
● Conclusion

All Roads Lead to Microservices
Cloud
aPaaS
Continuous Delivery
Dev Ops
Agile
Microservices

Microservices – What ?
The term "Microservice Architecture" has sprung up over the
last few years to describe a particular way of designing
software applications as suites of independently deployable
services.
-Martin Fowler
http://martinfowler.com/articles/microservices.html
Microservices are loosely coupled service oriented
architecture with bounded contexts.
- Adrian Cockroft

Monolith Application
● Imaginary E-Commerce App
● Very efficient inter process calls between
components
● Good architecture, until it needs to scale
during busy shopping season.

Monolith Application
● You want to scale only 'Shopping
Cart' and 'Checkout' Modules,
but instead have to scale the
entire App
● Other Challenges with multiple
database and Session
management

Monolith Application – Good Parts
● Contains Everything
● Single Deployment
● Minimum viable product can
be reached quickly
● Easy (or Familiar) to understand

Monolith Application – Bad Parts
● Complex Architecture
● Difficult to Scale
(cannot scale only selected
modules)
● Long term commitment
to one technical stack

Microservices
● Distillation of SOA
● Single Context
● Smart Services / Dumb Pipes
● RESTful Communications
● Smaller codebase – easy to
reason about
● Easy to scale

Challenges of distributed system
● Configuration Management
● Service Registration & Discovery
● Load Balancing
● Fault Tolerance
● Monitoring
● Concurrent API Integration & Transformation
*(This is not a complete list...)

Spring Boot
● Built on top of Spring framework
● Opinionated convention over configuration
● Creates stand-alone “Production” ready app
● Embedded Tomcat / Jetty
● Various Starters POMs to simplify configuration
● https://start.spring.io/ - Spring Initializr template project (We can
replace our kick start scripts with something like this)

Spring Boot
● It can get pretty small..
@RestController
class ThisWillActuallyRun {
@RequestMapping("/")
String home() {
"Hello World!"
}
}
Exercise 1 : Spring Boot Hello World

Spring Boot
● Retrieve data from database – Spring Data REST
● Entity Class
@Entity
@Table(name="users")
public class User implements Serializable{... }
● Repository Class
@Repository
public interface UserRepository extends JpaRepository<User, Integer> { }
● Main Class
@SpringBootApplication
public class SampleDataJpaApplication {
public static void main(String[] args) throws Exception {
SpringApplication.run(SampleDataJpaApplication.class, args);
}
}
● The same project runs in local and Cloud - PWS
Exercise 2 : Spring Data REST

Spring Cloud
● Writing a single service with Spring Boot is easy.
● But things gets complex with managing multiple services
● http://cloud.spring.io/spring-cloud-netflix/ , to the rescue.
● Similar to Spring Data umbrella project.

Spring Cloud Netflix
● Configuration Management – Spring Cloud Config
● Service Registration & Discovery – Eureka
● Load Balancing – Feign & Ribbon
● Fault Tolerance (Circuit Breaker) – Hysterix
● Monitoring – Hysterix Dashboard
● Concurrent API Integration & Transformation - RxJava

Spring Cloud Config
● Distributed configuration management.
● Provides server and client-side support for externalized configuration.
● Store Configuration properties (YAML content) in remote repository (Git
supported by default)
● Curl -X POST http://localhost:8888/bus/refresh, to get the latest values from
the config server.
Exercise 3 : Config Server
Cloud Bus (AMQP / RabbitMQ)
Git (default)
properties
Config Server
/refresh
/refresh

Service Registry & Discovery - Eureka
● Eureka is Netflix OSS Project for Service Discovery Server & Client.
● Eureka Server – can be configured to be highly available, supports server
replication for resiliency.
● Clients registers with Eureka and sends heartbeat messages to Server.
● Eureka Server
@EnableEurekaServer
class Eureka {
}
● Eureka Client
@EnableDiscoveryClient
public class Application {... }
● Producer Consumer example
Exercise 4 : Eureka – Producer / Consumer

Load balancing - Ribbon
● Ribbon is a Netflix OSS Project, it provides software load balancers with
cluster of servers.
● Provides basic functionality like supply the public DNS name of servers
to client, rotate among the list of servers according to certain logic.
● Load balancer components
Rule – logic to determine which server to return from the serverlist.
Ping – to ensure server liveness
ServerList – can be static or dynamic
● Spring RestTemplate has been added with little bit of magic and made
Ribbon enabled.
● FeignClient is alternative to Spring RestTemplate, Feign is declarative
web service client, makes writing web service clients easier.
Exercise 5 : Producer / Consumer example with load balancer / Feign

Fault Tolerance - Hystrix
● Hystrix – Circuit Breaker Pattern
● Its a state machine, with 3 states – Closed, Open and Half-Open.
● Failures does not cascade up to the top of the call stack.
Exercise 6 : Hystrix in action

Monitoring
● Hystrix Dashboard
Import org.springframework.cloud.netflix.hystrix.dashboard.EnableHystrixDashboard
@EnableHystrixDashboard
class HystrixDashboard { … }
Exercise 7 : Hystrix Dashboard in action

Concurrent API Integration - RxJava
● RxJava is Java implementation of MS
Reactive Extensions.
● A library for composing asynchronous
and event-based programs by using
Observable sequences.
● Netflix created RxJava to simplify
server side concurrency, a single
“heavy” client request that is executed
in parallel on the server.
● The service layer API method return an Observable<T> getData(), this can be
asynch or synch, and its transparent to the caller of this service.
● Can be achieved using Java Futures, but it supports one value, not for a
sequences. Also calling Future.get() will be a blocking.
● Callbacks can be used for asynchronous execution and works well for single level
of execution, but things gets complicated with nested Callbacks (callback hell !).

Conclusion
● Microservices Architectural Style needs certain infrastructure competencies
like, Rapid Provisioning, Continuous Delivery, Basic Monitoring, etc.,
● There are many new design patterns to learn when implementing a
distributed system.
● Spring Cloud Netflix implements certain patterns and Spring Boot makes it
easy to use.

References
● http://martinfowler.com/articles/microservices.html
● http://martinfowler.com/bliki/MicroservicePrerequisites.html
● http://projects.spring.io/spring-boot/
● http://projects.spring.io/spring-cloud/
● http://cloud.spring.io/spring-cloud-netflix/
● http://cloud.spring.io/spring-cloud-netflix/spring-cloud-netflix.html
● https://github.com/ReactiveX/RxJava
● http://www.infoq.com/presentations/reactive-arch-grails
● http://www.infoq.com/presentations/Netflix-API-rxjava-hystrix
Images
● https://www.flickr.com/photos/bhophoto/384574407/

Developing Microservices using Spring - Beginner's Guide

More Related Content

What's hot

Similar to Developing Microservices using Spring - Beginner's Guide

Recently uploaded

Developing Microservices using Spring - Beginner's Guide