DevOps and the Front-End Developer

Transcript

1. DevOps 
 and the 
 Front-End Developer Van Wilson @vanwilson Charlotte Front-End Developers Meetup Thursday, May 30, 2019
2. DevOps Mobile Responsive Agile Frameworks Pattern Library Data Viz Blockchain Machine
 Learning Augmented Reality Microservices
4. Does DevOps have 
 anything to do with 
 front-end development?
5. A Little History
6. The 1990s: 
 everyone is “full-stack” • designer/developer/server-admin is one and the same • aka “Webmaster” • "Deploying your code” means transferring it to the server with an FTP program
7. Even “better”, 
 edit code right on the server
8. As the web grew… …things changed
9. The March of Progress • 2005: Ajax, with its poster child, Google Maps • JavaScript libraries: Dojo, Prototype, YUI, and jQuery • 2008: GitHub • 2009: Node.js • 2010s: the age of frameworks, both CSS and JS
10. But wait, there’s more to the story
11. –Mike Loukides “From an operations standpoint, the big story of the web isn’t the evolution toward full-fledged applications that run in the browser; it’s the growth from single servers to tens of servers to hundreds, to thousands, to (in the case of Google or Facebook) millions.”
12. Goals for this talk 1. Give you an understanding of what DevOps is, what it isn’t, and why it matters to you. 2. Help you see the similarities between DevOps practices and common front-end development practices. 3. Prepare you to start, support, and advocate for DevOps in your organization.
13. What is DevOps?
14. –https://en.wikipedia.org/wiki/DevOps “DevOps is a set of software development practices that combines software development (Dev) and information technology operations (Ops) to shorten the systems development life cycle while delivering features, fixes, and updates frequently in close alignment with business objectives.” Definition
15. https://theagileadmin.com/what-is-devops/ “DevOps is the practice of operations and development engineers participating together in the entire service lifecycle, from design through the development process to production support.” More Definitionier
16. But why would those groups need to work together more closely?
17. The core conflict Goal of Devs:
 
 Respond to rapidly changing needs by frequently pushing new features
 Goal of Ops:
 
 Provide a stable, reliable, secure service
18. Illustration: Robin Johnston
19. Illustration: Jessica Wilson
20. Case Study:
 Knight Capital • Largest U.S. equities trader in 2012 • Had to launch new software on Aug. 1, 2012 to take advantage of new trading rules and maintain industry position • New software, SMARS, built on top of an old program, still deployed but no longer used
21. Case Study:
 Knight Capital (cont’d.) • Server updates done manually • Only 7 of the 8 servers were updated • No one remembered the old software was still on the 8th server • Order commands start going to all 8 servers at 8am on Aug. 1, 2012
22. Case Study:
 Knight Capital (cont’d.) • Issue not corrected for 45 minutes • Knight lost $440 million • Stock dropped 33% that day, 75% the next day • Taken over by a rival firm 6 months later
23. What DevOps Does • DevOps helps Developers get their code from their machines and into the hands of users more quickly, more consistently, and more reliably. • DevOps helps Operations people achieve levels of service with more uptime, more reliability, and more predictability.
24. The Roots of DevOps • Lean Manufacturing • Agile Software Development • The Toyota Production System
25. Value Streams https://www.lucidchart.com/pages/value-stream-mapping
26. Example Value Streams in Web Dev Design Develop Test Deploy Code Automated Testing Exploratory Testing Production Deploy
27. DevOps Helps to Answer these
 3 Questions about the Value Stream • How do I get the code I write in front of other people quickly? • How do I tell how well my code is working? • How do I learn from those first two experiences?
28. CAMS
 The Four Aspects of DevOps Culture Automation Measurement Sharing
29. Automation,
 aka Flow
30. Making Status Visible • Version control all the things, including your infrastructure • Automate your tests • Continuously integrate new work to the main branch • Automate your deployment and releases • Control what users see with “flags”
31. Version Control: Infrastructure as Code • Every sysadmin has always had a stash of scripts • Make all scripts visible to every one • Even database migrations • Use tools as needed: - Config mgt. (Puppet, Chef, Ansible, CFEngine) - Orchestration (Zookeeper, Mesos, Kubernetes) - Virtualization (AWS, Azure, Google Cloud) - Containerization (Vagrant, Docker)
32. Automated Testing • Unit tests • Integration tests • Functional tests • Code quality tests (Code Climate, Rubocop, Lint) • Security tests (Veracode)
34. Continuous Integration • Have a “build agent” run any automated tests • Whenever someone pushes a branch, or opens a PR • Create any build artifacts (compile, combine, minify) • Lots of options: Jenkins, Travis, CircleCI, TeamCity, Azure Pipelines
36. Automated Releases • The holy grail of DevOps • Good first step for your Dev environment • Next step, Test environment (bring QA on board) • Ship it straight to production! (what could go wrong?) • No more than goes wrong the old way, actually
37. Feature Flags • Checks added to the code where features are changing • On/Off status for each feature controlled externally • Business, or automation, can control new features without re-deploy • Tools can help: LaunchDarkly, Split, Rollout
38. Measurement,
 aka Feedback
39. Principles of Feedback • See problems as they occur • Swarm and resolves problems quickly (Andon cord) • Keep pushing quality closer to the source • Optimize for downstream work centers
40. Get Telemetry • Instrument your code • Many levels: business logic, application, environment • Tools: New Relic, Sentry, Rollbar, LogRocket, StatsD • You can even gather telemetry on your deployment pipeline
41. Analyze Telemetry • Don’t just log everything. Interpret. • Apply statistics. Means and standard deviations are easy wins. • Try to find the harbingers of problems (example, increased page load times) • Use tools to group problems • Want to see code quality skyrocket? Have devs take turns on call
42. Hypothesis-Driven Development • Don’t just rely on passive metrics. • Akin to “rapid prototyping” for UX research • Run experiments to see how things are working • Integrate A/B testing into the release process
43. Sharing,
 aka Learning
44. Myth of the Lone Genius
45. Enable Daily Learning • Take your metrics and testing, and apply it to improve • Netflix example: Chaos Monkey • Randomly terminates instances in production, to ensure that engineers implement their services to be resilient to instance failures • “Game days” - days designated to test for failures
46. Local Discoveries →
 Global Improvements • Chat rooms • Chat bots (e.g., GitHub’s Hubot) • Monorepo • Automated tests also serve as documentation
47. Time for Learning and Improvement • From Toyota: the kaizen blitz, a few days to concentrate on a problem area • Hack days, or hackathons • Google’s 20% time • Internal conferences • Internal coaching
48. Blameless Post Mortems • After an incident, a meeting to determine the situations and individual decisions that came together to cause the failure - construct a timeline - everyone relates their role in the situation - anyone who made a mistake is a teacher - propose countermeasures - everyone concerned attends
49. Culture
50. –Every person, in every industry, about every innovative idea, ever “This all sounds fantastic, and it may work for everyone else, but it doesn’t apply to 
 me / my company / our situation.”
51. Why does this not apply 
 to you? 1. I’m a freelancer. I work alone. 2. I work for a big enterprise company, in a heavily regulated industry. We would never be allowed to do this.
53. Enterprise companies that “do the DevOps” • Alcoa • Allstate • Amazon • Capital One • CSG • Etsy • Facebook • Google • HP • Intuit • LinkedIn • Nationwide • Netflix • Nordstrom • Pivotal Labs • Salesforce • Target • Twitter
54. –Matt Stratton “The first step of DevOps… is to find out how your organization makes money.”
55. Questions?
56. Resources General • “The evolution of DevOps”, Mike Loukides, https:// medium.com/@mikeloukides/the-evolution-of- devops-9d101124aa2e • “What Is DevOps?”, https://theagileadmin.com/about/ • The DevOps Handbook, Gene Kim, Jez Humble, Patrick DuBois, and John Willis, https://www.goodreads.com/ book/show/26083308-the-devops-handbook
57. Resources, cont’d Infrastructure as Code • “Infrastructure as Code: Everything You Need to Know”, Mark Robinson, https://rollout.io/blog/ infrastructure-as-code/ Continuous Integration • “Continuous Integration”, Martin Fowler, https:// www.goodreads.com/book/show/26083308-the-devops- handbook
58. Resources, cont’d Feature Flags • “Feature Toggles (aka Feature Flags)”, Pete Hodgson, https://rollout.io/blog/infrastructure-as-code/ Culture • Arrested Devops (podcast), Bridget Kromhout, Jessica Kerr, Matt Stratton, Trevor Hess, https:// www.arresteddevops.com/
59. Resources, cont’d Front-end specific DevOps • “Automating peace of mind with Accessibility Testing & Continuous Integration”, Marcy Sutton, https:// marcysutton.github.io/a11y-and-ci/#/