Scaling Engineering Organizations with Patterns
0 likes265 views
The document discusses patterns for scaling engineering organizations as they grow in size and complexity. It describes patterns for teams, feature teams, requirement areas, distributed architectures, microservices, platform teams, and developer meshes. Each pattern has pros and cons, and prerequisites are discussed for some patterns. The document emphasizes the importance of mastering best practices like continuous integration before scaling, and having leadership, a clear mission, and infrastructure to support more autonomous patterns.
Scaling Engineering Organizations with Patterns
- 1. Scaling Engineering Organizations with Patterns @sshishkin Sergey Shishkin
- 3. x1 x1 x1 WELCOME TO MUSHROOM INC.
- 4. Humble Origins • Unified product direction • Uncomplicated solution • Fast technical decisions
- 5. LEVEL 1
- 7. LET’S SCALE! x1 x1 x1
- 8. Pattern: Team • Pros: • Diverse ideas • Increased throughput • Decreased variability (bus factor) • Cons: • Cost of agreement • Contention of shared resources
- 9. Universal Scalability Law See: https://blog.acolyer.org/2015/04/29/applying-the-universal-scalability-law-to-organisations/
- 10. Best Practices • Collective Code Ownership • Coding Standard • Continuous Integration • Pair Programming • Test-Driven Development https://ronjeffries.com/xprog/what-is-extreme-programming/
- 11. Many organizations start to increase the number of teams before they master best practices with one team. This results in a bunch of underperforming teams and makes technical excellence harder to re-instill afterwards.
- 12. LEVEL 2
- 14. MORE TEAMS! x1 x1 x1
- 16. Pattern: Feature Teams • Pros: • Flexible prioritization of multiple stakeholders • Increased throughput • Cons: • Increased cost of coordination • Increased contention https://less.works/less/structure/feature-teams.html
- 17. LEVEL 3
- 20. Pattern: Requirement Area • Pros: • Manageable scope of prioritization • Customer focus • Potentially shippable backlog items • Cons: • Fragmented product vision • Cost of agreement https://less.works/less/less-huge/requirement-areas.html
- 21. LEVEL 4
- 24. Pattern: Distributed Architecture • Pros: • Partitioned domain complexity • Localized technology decisions • Localized testing • Cons: • Cost of agreement on service interfaces • Harder release coordination • New failure modes
- 25. Conway’s Law
- 26. LEVEL 5
- 27. CHALLENGE: GROWING DEMAND x15 x5 x10
- 29. Pattern: Micro-Services • Pros: • Reduced team contention • Scoped cognitive overhead • Cons: • Unpredictability of changes • Operations complexity
- 30. Pattern: Micro-Services https://martinfowler.com/bliki/MicroservicePrerequisites.html • Prerequisites: • Infrastructure as a service • Continuous deployment • Observability
- 31. https://blog.gardeviance.org/2015/03/on-pioneers-settlers-town-planners-and.html
- 33. Pattern: Platform Teams • Pros: • Unified solutions for common problems • Economy of scale • Competence development • Cons: • “One size fits all” solutions • Innovation bottleneck
- 34. Look out for existing bright spots and promote them
- 36. LEVEL 6
- 40. Pattern: Developer Mesh • Pros: • Motivating autonomy • Less organizational overhead • “Organic” prioritization • Cons: • No centralized product authority
- 41. Pattern: Developer Mesh • Prerequisites: • Clear company mission and purpose • Leadership at all levels
- 42. SUMMARY
- 45. GAME IS NOT OVER IT’S YOUR TURN! @sshishkin Sergey Shishkin
- 46. Links • Universal Scalability Law: https://blog.acolyer.org/2015/04/29/applying-the-universal-scalability-law-to- organisations/ • eXtreme Programming Practices: https://ronjeffries.com/xprog/what-is-extreme-programming/ • LeSS: Feature Teams: https://less.works/less/structure/feature-teams.html • LeSS: Requirement Areas: https://less.works/less/less-huge/requirement-areas.html • Conway’s Law: https://www.sketchplanations.com/post/175174474143/conways-law-im-paraphrasing- here-but-the-gist • Pioneers, Settlers, Town Planners: https://blog.gardeviance.org/2015/03/on-pioneers-settlers-town- planners-and.html • Dynamic Reteaming: http://www.heidihelfand.com/dynamic-reteaming/ • You have to be this tall to use Micro-Services: https://martinfowler.com/bliki/MicroservicePrerequisites.html • Cynefin Framework: https://www.agile42.com/en/agile-info-center/cynefin/ • How GitHub (no longer) works: https://www.infoq.com/presentations/github-evolution/
- 47. Image Credits • https://www.flaticon.com/authors/freepik • https://www.flaticon.com/authors/eucalyp • https://www.flaticon.com/authors/pixel-perfect • https://www.flaticon.com/authors/mynamepong • https://www.flaticon.com/authors/flat-icons • https://www.flaticon.com/authors/monkik • https://www.flaticon.com/authors/smashicons • https://findicons.com/icon/423523/paper_mario