[KCD UK] Platform Orchestrators: The missing middle of Internal Developer Platforms?

At previous KubeCons, I argued that developers building applications on Kubernetes today are being asked not just to code. They are also responsible for shipping and running their applications. We often talk about needing a Kubernetes “platform” to handle these responsibilities, but do we really need to build our own PaaS?

Key considerations I’ll cover include:

- Many organizations start their platform build with a developer portal, which provides success on day one but struggles with day two challenges.
- In many implementations, there appears to be a “missing middle” that manages the platform lifecycle, which a Platform Orchestrator can address.
- Most organizations are not as unique as they think they are, but they do need a way of codifying their business processes into the platform.

Transcript

1. @danielbryantuk Platform Orchestrators: The Missing Middle of Internal Developer Platforms? Daniel Bryant | Platform Engineer | Head of Product Marketing @ Syntasso @danielbryantuk
2. @danielbryantuk How do you build a platform? • Top-down, application developer-focused ◦ “The Backstage service catalog is fantastic, but the support for day 2 ops… not so muchˮ • Bottom-up, operations/infrastructure-focused ◦ “The Terraform workflow is fantastic, but infrastructure abstractions leak through to developers HCL, K8s, etc)ˮ • Middle-out, platform engineering-focused ◦ X-as-a-service, process automation, fleet management ◦ “Platform as a productˮ approach
3. @danielbryantuk @danielbryantuk (he/him) linktr.ee/danielbryantuk
4. @danielbryantuk Previously at
5. @danielbryantuk From Kubernetes to PaaS to… err, whatʼs next? https://www.youtube.com/watch?v=zUpYEhaUJnM
6. @danielbryantuk
7. @danielbryantuk What even is a platform? 🤔
8. @danielbryantuk What is a platform, anyway? 🚉 A digital platform is a foundation of self-service APIs, tools, services, knowledge and support which are arranged as a compelling internal product. Autonomous delivery teams can make use of the platform to deliver product features at a higher pace, with reduced coordination. Evan Bottcher martinfowler.com/articles/talk-about-platforms.html
9. @danielbryantuk Gartner: What is platform engineering? “Platform engineering improves developer experience and productivity by providing self-service capabilities with automated infrastructure operations. It is trending because of its promise to optimise the developer experience and accelerate product teamsʼ delivery of customer value.ˮ https://www.gartner.com/en/articles/what-is-platform-engineering
10. @danielbryantuk Gartner: What is platform engineering? “Platform engineering improves developer experience and productivity by providing self-service capabilities with automated infrastructure operations. It is trending because of its promise to optimise the developer experience and accelerate product teamsʼ delivery of customer value.ˮ https://www.gartner.com/en/articles/what-is-platform-engineering
11. @danielbryantuk Platform engineering expectations gartner.com/en/newsroom/press-releases/2023-11-28-gartner-hype-cycle-shows-ai-practices-and-platform-engineering-will-reach-mainstream-adoption-in-software-engineering-in-two-to-five-years
12. @danielbryantuk Platform engineering expectations linkedin.com/feed/update/urn:li:ugcPost:7199307746356199424
13. @danielbryantuk Platform challenges
14. @danielbryantuk The “whyˮ of platforms 🚉
15. @danielbryantuk Current software engineering challenges You want to overcome the following challenges: • Slow: Developers are blocked, waiting on tickets and other teams • Risky: Critical business processes are inconsistently applied • Inefficient: Tech sprawl leads to tech debt and operational overhead
16. @danielbryantuk Current software engineering challenges You want to overcome the following challenges: • Slow: Developers are blocked, waiting on tickets and other teams • Risky: Critical business processes are inconsistently applied • Inefficient: Tech sprawl leads to tech debt and operational overhead 50% of respondents are capable of deploying less than “between once per week and once per month” Accelerate State of DevOps 2023 42% of developers say they can’t release code to production without risking failures Harness State of the DecEx 2024 Overall, 59% of IT pros are frustrated by the time spent on routine infrastructure activities and 84% agreed that they “could bring more value to their organization if they spent less time on routine tasks” HorizonIQ Report 2020
17. @danielbryantuk
18. @danielbryantuk What are the goals of your platform? • Go faster: Platform teams need to provide “everything as a serviceˮ to help rapidly and sustainably deliver value to end-users • Decrease risk: Teams need to automate manual processes in reusable components • Increase efficiency: You need to manage and scale your digital platform and resources as a fleet
19. @danielbryantuk The “howˮ of platforms 🚉
20. @danielbryantuk Top down, app developer-focused rollout “Backstage is my platform. Developers go here to spin up a new application, deploy it, and view metricsˮ ✅ Fantastic developer experience (and service catalog) ✅ Highly customisable ❌ Often a facade that calls a series of infrastructure APIs ❌ Day 2 (automation) aspects of the portal and plugins can be challenging
21. @danielbryantuk “Terraform is my platform. I can orchestrate all of my infrastructure via HCL and cron jobs, and the GitOps pipelines automatically deploy applicationsˮ ✅ Everything-as-code ✅ Highly automatable ❌ Infrastructure abstractions leak outwards towards developers ❌ At scale, the diversity of tech can be challenging to automate/orchestrate Bottom up, operations-focused rollout
22. @danielbryantuk The missing middle? 🏗
23. @danielbryantuk https://thenewstack.io/crossplane-what-most-people-get-wrong-and-how-to-get-it-right/ https://cnoe.io/ https://platformengineering.org/platform-tooling https://www.kratix.io/
24. @danielbryantuk syntasso.io/post/platform-engineering-orchestrating-applications-platforms-and-infrastructure
25. @danielbryantuk syntasso.io/post/platform-engineering-orchestrating-applications-platforms-and-infrastructure
26. @danielbryantuk Gartner: What is platform engineering? “Platform engineering improves developer experience and productivity by providing self-service capabilities with automated infrastructure operations. It is trending because of its promise to optimise the developer experience and accelerate product teamsʼ delivery of customer value.ˮ https://www.gartner.com/en/articles/what-is-platform-engineering
27. @danielbryantuk Making it happen 🚉
28. @danielbryantuk Platforms: Abstract thinking • APIs, domains, SOLID/CUPID ◦ Just as important for platforms ◦ Think information hiding ◦ Build for progressive disclosure • Design for UI, CLI, & API ◦ Meet devs where they are at ◦ Optimise for automation syntasso.io/post/when-backstage-met-terraform-a nd-platform-orchestrators-webinar-recap
29. @danielbryantuk Business requirements change, user expectations expand, APIs are at the core of sustainable platforms • Open Application Model (OAM) / Score • Kratix and Promises • KubeBuilder • Crossplane • Massdriver • … More user focused More implementation focused Platform abstractions (and APIs)
30. @danielbryantuk Tech "stacks" are emerging • The “BACKˮ stack: Backstage, Argo, Crossplane, Kyverno • CNOE Framework: Cloud Native Operation Excellence • Kubefirst • DIY All the other CNCF tech++ • … More opinionated Less opinionated Platform automation (tech)
31. @danielbryantuk This can be overwhelming…
32. @danielbryantuk Example: Kratix
33. @danielbryantuk Promises, promises…
34. @danielbryantuk
35. @danielbryantuk Build it, and they will come Solve problems, and they will come
36. @danielbryantuk What are the goals of your platform? • Go faster: Platform teams need to provide “everything as a serviceˮ to help rapidly and sustainably deliver value to end-users • Decrease risk: Teams need to automate manual processes in reusable components • Increase efficiency: You need to manage and scale your digital platform and resources as a fleet
37. @danielbryantuk Product focus thoughtworks.com/en-gb/insights/looking-glass/platforms-as-products
38. @danielbryantuk Product focus thoughtworks.com/en-gb/insights/looking-glass/platforms-as-products
39. @danielbryantuk What gets measured, gets managed • Establish goals and measurements • Leading indicators ◦ Adoption rates ◦ Onboarding times ◦ Time to nth PR • Lagging indicators ◦ App retention rate ◦ Incidents/near misses mitigated ◦ ROI (time saved)
40. @danielbryantuk Wrapping up 🎉
41. @danielbryantuk Conclusion • 🏗 Build your platform intentionally • 🚉 Watch out for the missing middle: “platform orchestratorsˮ • 👀 Struggling with scaling day 2 operations with your portal? • 👀 Infrastructure abstractions leaking to developers? • 🎯 Focus on “platform as a productˮ
42. @danielbryantuk Thank you! @danielbryantuk syntasso.io/post/platform-engineering-orchestrating-applications-platforms-and-infrastructure docs.kratix.io/main/quick-start speakerdeck.com/danielbryantuk slideshare.net/dbryant_uk/presentations