"Go big or go Home!" The Beauty of the Monolith

There probably has not been a bigger buzzword creating waves in development circles quite as the idea of Microservices did over the last 10 years. Netflix somehow made this work and now every ambitious developer is slaying perfectly good monoliths into tiny pieces just to catch a ride on the bandwagon. In his talk, Philipp will explain why Microservices will neither magically erase architecture flaws nor increase the amount of work packages that can be done in parallel. Contrary, the concept rather serves as a catalyst amplifying many of the problems developers were hoping to get rid of. While inexplicably gone out of fashion, Philipp will explain why the monolith still has its right to exist to this day. As Public Enemy said it best back then in 1988: "Don't Believe the Hype".

Transcript

3. 3 WHAT ABOUT ME? • Developer for 7+ years • @ Posedio since 2022 • Sagittarius with Virgo rising • Cat Lover • Informatics as a craft not a science Philipp Seifert-Kehrer
4. ROADMAP 1. Xs and Os 2. Where Promised Land? 3. It’s Tricky getting it right 4.The Majestic Monolith 5. Pretty sus Standard
5. MICROSERVICES – MONOLITHS XS AND OS 1
6. 6 HOW COME I GATHERED YOU HERE TODAY
7. 7 MICROSERVICES? MONOLITHS? Many small applications Do 1 thing well Communication via HTTP Teams own specific services Wut? One deployable
8. 8 ADVERTISED BENEFITS Easier to build and maintain Improved productivity and speed Flexibility in technologies Autonomous teams Improved Scalability and Fault Isolation
9. 9 SO, WHAT TO EXPECT FROM THIS TALK? Questioning microservices as standard architectural approach in modern applications AND Arguing for monoliths as a solid architectural choice in modern applications
11. 11 RESPONSIBILITY PING PONG
12. 12 PIPE DREAM OF FASTER DEVELOPMENT
13. 13 INDEPENDENCE We can change implementation without coordinating with others
14. 14 MICRO… IS PRETTY DAMN SMALL Starting off too small Frontend glue How small is too small?
15. 15 AMPLIFIED PROBLEMS Increased Load/Latency Fault Isolation only technically More possible sources of error
17. IT’S TRICKY 3
18. 18 EVERYONE CAN BE A DEVELOPER
19. 19 DISTRIBUTED SYSTEM
20. 20 GAMEPLAN Single team or outstanding communication Profound knowledge of domain Clear-cut understanding of non-functional requirements Developers that are skilled and accountable
21. 21 BUT… … unless you really really really have to!!!
23. THE MAJESTIC MONOLITH 4
24. 24 LEGACY UP YOURS
25. 25 MONOLITHS ARE NOT PERFECT, BUT… Easy to refactor
26. 26 WHAT YOU WANT What you really really want
27. 27 SEPARATION OF CONCERNS https://github.com/spring-projects/spring-modulith
28. 28 THAT FANCY SPRING PROJECT
29. A PRETTY STANDARD 5
30. 30 GETTING IT RIGHT FROM THE GET GO
31. 31 ADMIT YOU DON’T KNOW SHIT… … at the beginning of a project/startup/whatever… … but are willing to learn and adapt accordingly.
32. 32 DON’T BE THAT GUY/GAL
33. 33 WHO BENEFITS MOST FROM MICROSERVICE HYPE?
34. 34 NICE CONCEPT Reality hits different Solution for very specific problems New problems to solve
35. 35 BEST PART ABOUT MICROSERVICES? They tend to be easily refactored into a monolith.
36. 36 WRAPPING IT UP Maintainability Availability Scalability Sustainability Get away from microservices as a standard Stop trying to outsmart the world Start off with a monolith Adapt according to the problems at hand Go big or go home
37. 37 “So, don’t be a prick Admit you don’t know shit And start off any project with some nice monolith” - Drake
38. THANK YOU! POSEDIO GMBH +43 660 9154419 contact@posedio.com Weyringergasse 1-3, 1040 Wien, Millenium Park 4, 6980 Lustenau www.posedio.com Be careful on the cold concrete!