Not a SO(A) Trivial Question!

MuCon presentation about the legacy of SOA and its influence on Microservices today.

Transcript

1. A Not SO(A) Trivial Question! Tareq Abedrabbo - MuCon 7/11/2017
2. Agenda • Motivation • SOA vs Microservices • Design: anti-patterns and recommendations
3. About Me • CEO/Consultant at OpenCredo • Background • SOA • Spring-WS • Microservices
4. Motivation
5. Background • Some organisations are starting greenfield microservices projects • A few are trying to evolve preexisting platforms • Many of these organisations have a SOA heritage
6. Microservices = SOA done right ?
7. What is SOA?
8. “A service has four properties according to one of many definitions of SOA: 1.It logically represents a business activity with a specified outcome. 2. It is self-contained. 3. It is a black box for its consumers. 4. It may consist of other underlying services.”
10. In reality SOA is also….
11. ➡ An architectural style and design patterns ➡ Data models and patterns ➡ Technology and tools (ESBs, GUIs) ➡ A collection of specifications (SOAP, WS-*)
12. The SOA Heritage ✓ Approach & design ✓ Technology selection ✓ Team organisation
13. “Those who cannot remember the past are condemned to repeat it”
14. SOA vs Microservices
15. Reuse vs Managing change
16. Vendor-driven view vs Broad system view
17. Integration vs Composition
18. Technical reuse vs Functional reuse
19. Static vs Dynamic
20. Organisational silos vs Collaborative teams
21. Design
22. Some anti-patterns
23. Example 1: The Distributed Monolith
24. Example 2: The Revenge of the ESB
25. • Using ESBs (or similar) • Ignoring delivery guarantees/expecting transactions • Excessive marshalling/unmarshalling • Canonical data models/complex schema • Distributed monolith (based on lack of understanding of boundaries) • Minilith (proliferation of technology, databases) • lack of automation in infrastructure/testing
26. Recommendation • No canonical data model (services do not need to understand every aspect of the service) • canonical data model vs schemas • Separate data from meta-data • Data orientated modelling/schemas • Normalise meta-data
27. Some Recommendations
28. Keep it domain-driven
29. Replace canonical data models with data views
30. Normalise metadata (and separate it from the rest of the data)
31. Use the right tool for the job (and use common sense!)
32. Thank you!