2024-06 ICSA Industrial Keynote - Demystifying Hypes in Software Architecture
0 likes80 views
The document discusses the evolution and adoption of cloud-native technologies, particularly the shift from microservices to serverless computing (FaaS) due to benefits like reduced operational overhead and automatic scaling. It highlights the motivations for migration, the complexities and challenges faced, and introduces concepts like micro-frontends and architectural drift analysis. Ultimately, it emphasizes that while new technologies can offer advantages, careful consideration of associated issues and architectural choices is crucial.
![What is Serverless [3]
[3] Baldini I. et al. (2017) Serverless Computing: Current Trends and Open Problems. In: Chaudhary S., Somani G., Buyya
R. (eds) Research Advances in Cloud Computing. Springer, Singapore](https://image.slidesharecdn.com/2024-06icsakeynote-demistifyinghypes-240930160057-5dffc3a9/85/2024-06-ICSA-Industrial-Keynote-Demystifying-Hypes-in-Software-Architecture-10-320.jpg)
![• 19 Research-based tools for Architectural
reconstruction [Bakhtin et al]
• Arcan (Univ. Milano Bicocca)
• Prophet (Oulu and Baylor University)
• …
Static Analysis for Microservice-Based Applications
https://github.com/cloudhubs/prophet-web
Static Analysis: Tools
Alexander Bakhtin, Xiaozhou Li, Jacopo Soldani, Antonio Brogi, Tomas Cerny, and Davide Taibi. 2023.
Tools Reconstructing Microservice Architecture: A Systematic Mapping Study. AMC@ECSA Workshop](https://image.slidesharecdn.com/2024-06icsakeynote-demistifyinghypes-240930160057-5dffc3a9/85/2024-06-ICSA-Industrial-Keynote-Demystifying-Hypes-in-Software-Architecture-38-320.jpg)
![18 tools
(10 commercial, 8 research-based)
[Bakhtin et al]
Dynamc Analysis: Tools
Alexander Bakhtin, Xiaozhou Li, Jacopo Soldani, Antonio Brogi, Tomas Cerny, and Davide Taibi. 2
Tools Reconstructing Microservice Architecture: A Systematic Mapping Study. AMC@ECSA Work
Jaeger
Zipkin
AWS X-Ray
Netflix Interactive VIsualization](https://image.slidesharecdn.com/2024-06icsakeynote-demistifyinghypes-240930160057-5dffc3a9/85/2024-06-ICSA-Industrial-Keynote-Demystifying-Hypes-in-Software-Architecture-41-320.jpg)
2024-06 ICSA Industrial Keynote - Demystifying Hypes in Software Architecture
- 1. Demystifying Hypes in Cloud- Native Technologies Davide Taibi Professor. University of Oulu
- 3. 2015s Function as a Service Highly Decoupled Software Evolution: Organization
- 6. Why Practitioners Adopt New Technologies
- 7. Microservice Migration Motivations - Improve Maintainability - Improve Scalability - Reduce Costs - Simplify Distributed Work - Improve System understandability
- 8. Main Issues Identified Davide Taibi - Colloquium Series - SPLAB ZHAW Technical Issues • Architectural Patterns are not clear • Decoupling from the monolithic system • Database migration and data splitting Effort-Related issues • Effort required for the DevOps infrastructure • Effort required for library conversion • Effort at least 20% higher
- 9. O’Reilly SW Architecture Conference 2018 Stop using microservices! Move to serverless functions as soon as possible!
- 10. What is Serverless [3] [3] Baldini I. et al. (2017) Serverless Computing: Current Trends and Open Problems. In: Chaudhary S., Somani G., Buyya R. (eds) Research Advances in Cloud Computing. Springer, Singapore
- 11. From Microservices to Serverless Functions • Practitioners started migrating from microservices to FaaS • Mixed Approach (microservices + Functions) • New applications 100% serverless
- 12. Why practitioners are moving to serverless
- 13. Migration Motivations Companies Already moved to Microservices • OPS Effort for Microservices • Get rid of Kubernetes • No OPS Companies Migrating from Monolithic systems • New (hype) technology • Promising technology • No initial infrastructural costs (pay as you use) • Automatic scaling • Lack of skilled OPS personnel
- 14. Migration Issues • Developers are not used to the event-oriented programming • Very hard to test • Debug almost impossible • Unknown Patterns and antipatterns • Anomalies can generate unexpected costs!
- 16. Both! Serverless Microservices Composing Serverless functions to Microservices
- 18. “ Micro-frontends are the technical representation of a business subdomain, they allow independent implementations with same or different technology choices, finally they avoid sharing logic with other subdomains and they are own by a single team 21 From Domain Driven Design (DDD)
- 19. 2
- 20. How About Service Mesh?
- 22. Companies are rolling back R.Su, X.Li, D.Taibi. Back to the Future: From Microservice to Monolith. Microservices 2023, Pisa.
- 23. From Microservice to Monolithic Microservices Serverless Micro- Frontends Cognitive and Continuum Cloud
- 24. From Microservice to Monolithic Microservices Serverless Micro- Frontends Cognitive and Continuum Cloud Reasons: complexity, scalability, performance, organizational reasons.
- 26. Benefits and Issues of new Technologies Patterns and Anti-Patterns Follow-Up Case Studies Technology Assessment Frameworks Our Demystification Process
- 27. Demystifying New Hypes Microservices Serverless Micro- Frontends Cognitive and Continuum Cloud
- 29. Technology Assessment Feramework • Microservice Assessment Framework
- 30. Technology Assessment My favourite example (Blockchain) M.E. Pech DO YOU NEED A BLOCKCHAIN? IEEE Spectrum. September 2017. https://spectrum.ieee.org/do-you-need-a-blockchain
- 31. Architectural Drift Analysis and Degradation Long-Term Architectural Choices Shouldn’t be violated
- 32. Architectural Drift Analysis and Degradation Is your architecture changed?
- 33. Architectural Drift Analysis and Degradation Is your architecture changed? Reverse Engineering – Architectural Reconstruction
- 34. Architectural Reconstruction • Static • From source code • Dynamic • From the execution of the system (e.g. log traces)
- 35. Mapping all the possible “paths” in your system 38 Static Analysis
- 36. 39 Credits: Google Maps Static Analysis
- 37. • Source Code • IaC • Git Log • Commit message • PR • Comments • Issue Trackers. • … 40 Static Analysis
- 38. • 19 Research-based tools for Architectural reconstruction [Bakhtin et al] • Arcan (Univ. Milano Bicocca) • Prophet (Oulu and Baylor University) • … Static Analysis for Microservice-Based Applications https://github.com/cloudhubs/prophet-web Static Analysis: Tools Alexander Bakhtin, Xiaozhou Li, Jacopo Soldani, Antonio Brogi, Tomas Cerny, and Davide Taibi. 2023. Tools Reconstructing Microservice Architecture: A Systematic Mapping Study. AMC@ECSA Workshop
- 40. 43 Dynamic Analysis. Street Traffic View Bicycle Traffic View
- 41. 18 tools (10 commercial, 8 research-based) [Bakhtin et al] Dynamc Analysis: Tools Alexander Bakhtin, Xiaozhou Li, Jacopo Soldani, Antonio Brogi, Tomas Cerny, and Davide Taibi. 2 Tools Reconstructing Microservice Architecture: A Systematic Mapping Study. AMC@ECSA Work Jaeger Zipkin AWS X-Ray Netflix Interactive VIsualization
- 42. Existing tools shortfalls • Reconstruction only • Mainly for visualization • High potential for SQA
- 43. What can we measure from Call Graphs?
- 44. Architectural Quality 4 - Microservice Patterns and Anti-Patterns - Identification - Detection tools and methods Anti-Patterns Detection and Visualization
- 45. 4 - Structural Coupling - Logical Coupling - Organizational/Architectural Coupling Microservice Coupling Visualization Example Microservice Coupling and Anti-Pattern Visualization Example Architectural Quality: Coupling
- 46. A possible future… with generative AI Refactoring and Quality Improvement • Will it clean up our code? • Will it refactor our SW Architecture? • Will it enable to remove simple anti-patterns or prevent them?
- 47. Conclusion • New Technologies are continuously introduced • Silver Bullets do not exist • Economical, and Technical motivations should be always considered • Think about the issues in your system before migrating
- 48. Thanks Davide Taibi University of Oulu davide.taibi@oulu.fi https://m3s-cloud.github.io