Lightweight Java EE for <Buzzword> Services @ JBCNconf 17

June 19, 2017 @ JBCNconf 17

Java EE is not a hype. It is a well known player since 1999 – a quite long time but Java EE is just about to come off age.

“Modern” developers create micro or nano services, they use “state of the art” and “lightweight” frameworks. But when is a service micro or nano? When is a framework lightweight? How does Java EE fit in this modern world of software development?

In my opinion Java EE is a perfect choice for writing micro or nano services. In this talk I try to prove my statement. Live on stage I write two services with RESTful JSON APIs with only a few lines of code which communicate with each other and run the services using a local Docker cloud where you can see how these can be scaled up and down to manage fluctuating loads. Coding, building, deploying, scaling: fast and efficient!

To fully understand this talk the attendees should have basic knowledge of Java.

Transcript

1. Lightweight Java EE for <Buzzword> Services Coding, building, deploying, scaling: fast and efficient! Marcus Fihlon, McPringle June 19, 2017 Software Engineer | Agile Coach | Lecturer | Speaker | Author
2. Disclaimer The following presentation has been approved for open audiences only. Hypersensitivity to occasional profanity requires covering ears. All logos, photos etc. used in this presentation are the property of their respective copyright owners and are used here for educational purposes only. Any and all marks used throughout this presentation are trademarks of their respective owners. The presenter is not acting on behalf of CSS Insurance, neither as an official agent nor representative. The views expressed are those solely of the presenter. Marcus Fihlon disclaims all responsibility for any loss or damage which any person may suffer from reliance on this information or any opinion, conclusion or recommendation in this presentation whether the loss or damage is caused by any fault or negligence on the part of presenter or otherwise. 1
3. Session Material Slides, Code, Video http://fihlon.ch/jbcnconf17 2
4. About Me Software Engineer CSS Insurance, Open Source Software Agile Coach CSS Insurance Lecturer TEKO Swiss Technical College Speaker Conferences, User Groups, Meetups Author Articles, Books www.fihlon.ch | github.com | hackergarten.net | jug.ch 3
5. Agenda Buzzword Bingo Theory of Evolution Live Coding Wrap-up 4
6. Buzzword Bingo
7. Buzzword Bingo Monolith Bad by default! Microservices Solve every problem! Nanoservices Solve all other problems! Docker Just because it’s cool! Java EE Uncool and heavy framework for bloated monoliths! 5
8. Buzzword Bingo Monolith Bad by default! Microservices Solve every problem! Nanoservices Solve all other problems! Docker Just because it’s cool! Java EE Uncool and heavy framework for bloated monoliths! 5
9. Buzzword Bingo Monolith Bad by default! Microservices Solve every problem! Nanoservices Solve all other problems! Docker Just because it’s cool! Java EE Uncool and heavy framework for bloated monoliths! 5
10. Buzzword Bingo Monolith Bad by default! Microservices Solve every problem! Nanoservices Solve all other problems! Docker Just because it’s cool! Java EE Uncool and heavy framework for bloated monoliths! 5
11. Buzzword Bingo Monolith Bad by default! Microservices Solve every problem! Nanoservices Solve all other problems! Docker Just because it’s cool! Java EE Uncool and heavy framework for bloated monoliths! 5
12. Theory of Evolution
13. Expensive Resources 6
14. Add Virtualization 7
15. One VM per application 8
16. Shared Resources with Docker 9
17. Microservice Hype 10
18. Fat Binaries: long build times 11
19. Use the layers: short build times 12
20. Keep it lightweight! 13
21. Live Coding
22. Architecture 14
23. Prepare the Swarm List all available machines docker-machine ls Create a new machine as a manager docker-machine create -d virtualbox mgr Create a new machine as a worker docker-machine create -d virtualbox node01 Login to the manager docker-machine ssh mgr Initialize the swarm docker swarm init \ --advertise-addr 192.168.99.100 15
24. Prepare the Visualizer Start the visualizer service docker run -it -d -p 8080:8080 \ -e HOST=192.168.99.100 -v \ /var/run/docker.sock:/var/run/docker.sock \ manomarks/visualizer Open the visualizer service in a web browser http://192.168.99.100:8080/ 16
25. Join the Swarm Ask the manager node for the token to join the swarm docker swarm join-token worker Login to the worker node docker-machine ssh node01 Join the worker to the swarm docker swarm join --token <token> \ 192.168.99.100:237 17
26. Verify the Swarm Login to the manager docker-machine ssh mgr List all nodes docker node ls Show more detailed information docker info 18
27. Create a Network Login to the manager docker-machine ssh mgr Create a new overlay network docker network create -d overlay jbcnconf List all networks docker network ls 19
28. Deploy Services Login to the manager docker-machine ssh mgr Deploy the time service docker service create --network jbcnconf \ --name timeservice mcpringle/timeservice Deploy the hello service docker service create --network jbcnconf \ -p 8181:8080 \ --name helloservice mcpringle/helloservice List all services docker service ls 20
29. Testing, Scaling and Draining Testing our services curl \ http://192.168.99.100:8181/api/hello/jbcnconf Scaling services up and down docker service update \ --replicas 3 helloservice Draining the manager node docker node update --availability drain mgr 21
30. Wrap-up
31. Conclusion With Java EE and Docker you get… easy to understand code fast build times reproducable deployments easy scaling of your services You are fast and efficient because you have your focus on your business code. You create business value! 22
32. Thank You! Questions? Slides, Code, Video http://fihlon.ch/jbcnconf17 23