Developer Velocity
- 2. Chris Reddington Cloud Solution Architect at Microsoft UK Blogger, Podcaster & Vlogger @ www.cloudwithchris.com Consider a follow or subscribe! Twitter: https://twitter.com/reddobowen YouTube: https://youtube.com/c/CloudWithChris GitHub: https://github.com/chrisreddington Steph Locke Cloud Solution Architect Manager at Microsoft UK Ex-Data&AI MVP Blog: https://stephlocke.info Twitter: https://twitter.com/theStephLocke LinkedIn: https://linkedin.com/in/StephanieLocke GitHub: https://github.com/stephlocke
- 7. Key findings Top drivers for business performance: developer tools, product management capabilities, culture, and talent management Open source is the #1 differentiator for top performers Accelerating impact: AI/ML, Live site automation, product telemetry Public cloud adoption is a key differentiatordriving business performance for non-digital companies Top performers: empowering citizen developers score 33% higher on innovation </>
- 13. An evolving journey of Developer Velocity Buildingthe Foundation Transforming the Future Accelerating the Impact Talent management Product management capabilities Automated live site management Public cloud adoption Effective developer toolchain Open source adoption and capabilities AI / ML Culture of psychological safety and ‘fail fast & learn’ Empowering citizen developers Usage of granular product telemetry </> AI
- 14. 65% 47%
- 15. #1 differentiator 30% 20% 4x higher 33% </>
- 16. AI
- 17. What is DevOps? Development + operations DevOps is the union of people, processes, and technologies to deliver continuous value to users.
- 18. Here’s how Microsoft builds apps of the future Adopting a growth mindset Creating a culture of customer obsession Aligning to shared objectives
- 19. DevOps across Microsoft http://aka.ms/DevOps-Stories 105K Engineers use the DevOps platform 2M Git commits per month 4.4M Builds per month 500M Test executions per day 5M Work items viewed per day 500K Work items updated per day 85,000 Deployments per day Internal Microsoft Engineering System Activity, July 2019
- 20. Create clarity Measure outcomes not outputs Objectives and key results (OKRs) Example objective: Grow a strong and happy customer base 1. Increase external NPS from 21 to 35 2. Increase docs SAT from 55 to 65 3. New pipeline flow has an Apdex score of 0.9 4. Queue time for jobs is 5 seconds or less KRs are measures for the quarter. Best KRs are leading indicators. Encourage ambitious KRs: 70% of the improvement target scores green.
- 21. Create clarity Change what you track Focus on measuring only the most critical and impactful KPIs: Customer usage How much value are users getting? Acquisition Retention Engagement Satisfaction (NPS) Feature usage Pipeline throughput How efficient is the DevOps process? Time to build Time to test Time to deploy Time to improve Failed and flaky automation Live-site health How quickly can you detect and fix issues? Time to detect, time to communicate, time to mitigate Customer impact, customer support metrics Incident prevention items Aging live-site problems SLA per customer
- 22. Be customer obsessed Ensure products are live in production and collecting feedback Collect telemetry data that examines the hypothesis that motivated the deployment. Gather information and making incremental changes is key to improvement. Treating the goal as a question rather than a statement of fact motivates developers to continue testing their ideas.
- 23. Evolve to full stack teams Evolve the organization (1 of 3) Program management Development Testing
- 24. Evolve to full stack teams Evolve the organization (2 of 3) Program management Engineering Ops/SRE
- 25. Evolve to full stack teams Evolve the organization (3 of 3) Feature team
- 28. Change how you collaborate, develop, and deliver Collaborate on code Leverage innersource practices to build better products and create a culture of openness by: Building upon open-source projects that use global contributors Customizing, adapting, and reusing existing internal code Collaborating with developers across your organization Appointing a smaller team to make final decisions Leveraging open- source software and innersource practices enables your team to innovate and deliver value faster.
- 29. Change how you collaborate, develop, and deliver Drive toward automation Long running functional tests Shifted to unit tests from automated functional tests Count of tests M62 M63 M64 M65 M66 M67 M68 M69 M70 M71 M72 M73 M74 M75 M76 M77 M78 M79 M80 M81 M82 M83 M84 M85 M86 M87 M88 M89 M90 M91 M92 M93 M94 M95 M96 M97 M98 M99
- 30. Change how you collaborate, develop, and deliver Utilize pull requests Control changes Pull requests are the first gate to control how changes get merged to the master. Limit human error and ensure only high-quality code gets merged with detailed permissions, policies, and status checks.
- 31. Enhance security from dev to production Secure the software supply chain Apply policies and controls proactively Rely on security automation Build security checks into your CI/CD pipelines Strengthen your security posture
- 32. Enhance security from dev to production Build security practices into your delivery Key security practices: Pre-commit • Threatmodeling • IDE securityplug-in • Pre-commithooks • Secure coding standards • Peer review Commit (CI) • Static code analysis • Securityunit tests • Dependency management Deploy (CD) • Infra as code (IaC) • Securityscanning • Cloud configuration • Securityacceptance tests Analyze • Security Smoke tests • Configuration checks • Penetration tests Operate • Continuous monitoring • Threatintelligence • Penetration testing • Blameless postmortems
- 33. An evolving journey of Developer Velocity Buildingthe Foundation Transforming the Future Accelerating the Impact Talent management Product management capabilities Automated live site management Public cloud adoption Effective developer toolchain Open source adoption and capabilities AI / ML Culture of psychological safety and ‘fail fast & learn’ Empowering citizen developers Usage of granular product telemetry </> AI