Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Agile & the Large Organization: A Perfect Match

Agile & the Large Organization: A Perfect Match

Deliver working, bug-free software daily using Acceptance Test Driven Development, TDD, Continuous Integration, Automated Infrastructure and Cloud Based Deployments connected to Internal Data Centers. All of this in a regulated, siloed, schedule driven and extremely political environment. Add to this long release cycles, slow hardware procurements, technology review boards, and cranky sysadmins. Think it's not possible?

Join Alyssa Riedl, Director at the US Department of Treasury and Nayan Hajratwala, Lean/Agile Coach as they discuss their experiences delivering a multi-billion dollar project based on Agile Principles and Practices where no one thought it was possible and without the use of any branded "Scaling Frameworks". Spoiler Alert: Requires a commitment to effective collaboration and an open culture.

Nayan Hajratwala

July 31, 2014
Tweet

More Decks by Nayan Hajratwala

Other Decks in Technology

Transcript

  1. A G I L E & T H E L

    A R G E O R G A N I Z AT I O N : N A YA N H A J R A T WA L A C O A C H - K C P C S P C S D C S M C J P N A YA N @ C H I K L I . C O M T W I T T E R : @ N H A J R A T W A P E R F E C T M A T C H A LY S S A R I E D L D I R E C T O R D E B T M A N A G E M E N T S E R V I C E S A LY S S A . R I E D L @ F I S C A L . T R E A S U RY. G O V
  2. P E O P L E P R O B

    L E M S • Command & Control Leadership • Changing Team Dynamics • Siloed Skill Sets • Shared Operations Staff
  3. P R O C E S S P R O

    B L E M S • Long Release Cycles • Slow Hardware Procurement • Technology Review Board • Security Reviews • Change Management
  4. L O N G R E L E A S

    E C Y C L E S P R O C E S S P R O B L E M # 1
  5. • 6 months to get a normal, low-risk change into

    production due to gates & manual testing, but…
  6. Build Artifacts + Isolation Tests Automated Acceptance Tests! (Cucumber, Selenium)

    Pushbutton Deploy AWS Smoke Test Source Control Automated Security & Quality Scans Pushbutton Deploy Internal Data Center Smoke Test Production! Build Environment! (Chef)
  7. S E T U P F O R C O

    N T I N U O U S D E L I V E RY • Team Co-location • Small Features • Three Amigos (ATDD) • Continuous Integration • Pushbutton Deployments • Automate Everything
  8. S L O W H A R D WA R

    E P R O C U R E M E N T P R O C E S S P R O B L E M # 2
  9. B E F O R E 1. Write business case

    - 1 day 2. Wait for approval - 1 week 3. Order Server - 1 day 4. Wait for delivery - 1 week 5. Install Server in Rack - 1 week 6. Install Software - 2-3 weeks Total: 6+ weeks
  10. A F T E R 1. Decide I need a

    server - 5 minutes 2. Log in to AWS and create server - 5 minutes 3. Wait for server to start (5 minutes) 4. Run Chef Cookbooks (5 minutes - 2 hours) Total: < 3 hours
  11. S T I L L N E E D S

    W O R K • Production deployments still occurring in internal data center… • So new production servers still have a long lead time
  12. T E C H N O L O G Y

    R E V I E W B O A R D P R O C E S S P R O B L E M # 3
  13. • Become familiar with the process • Quickly submit any

    new desired technologies • If it meets a business need, use it. • Never deselect technology because it “isn’t on the list”
  14. S E C U R I T Y R E

    V I E W S P R O C E S S P R O B L E M # 4
  15. B E F O R E • Code: • Team

    level code reviews • Pre-deployment scans • DB: • Manual updates by DBAs • Server • Manual updates by SAs
  16. A F T E R • Code: • Paired Programming

    • Scans integrated into CI • DB: • Versioning & Automation (Flyway) • Server • Versioning & Automation (Chef)
  17. S T I L L N E E D S

    W O R K • Paired Programming not occurring 100% of time • Database replication causing some issues with automation • Internal Data Center has some manual work to create servers
  18. C H A N G E M A N A

    G E M E N T P R O C E S S P R O B L E M # 5
  19. • Address root regulations rather than accept standard practices •

    Build process around best & most lightweight tools possible
  20. P E O P L E P R O B

    L E M S • Command & Control Leadership • Changing Team Dynamics • Siloed Skill Sets • Shared Operations Staff
  21. C O M M A N D & C O

    N T R O L L E A D E R S H I P P E O P L E P R O B L E M # 1
  22. • Problem: Repeated disappointments lead to lack of trust in

    teams to deliver. • Solution: Demos every 2-3 weeks showing running, tested, features. No Mockups!
  23. • Problem: Desire to build “ilities” framework - security, configurability,

    scalability • Solution: Define specific examples of these in the context of a feature using the INVEST mnemonic
  24. I N V E S T • I - Independent:

    is self contained • N - Negotiable: can be changed until started • V - Valuable: for the end user • E - Estimable: have some idea of how big it is • S - Small: no more than 3 days development effort • T - Testable: must know whether it works. (automate!)
  25. • Problem: Desire to develop a process that can be

    dropped in to other teams • Solution: Continuous education about the core principles of agile • http://j.mp/nayan-agile-principles • https://speakerdeck.com/nhajratw/agile-from- principles-to-practice
  26. I T ’ S A B O U T T

    H E J O U R N E Y
  27. • Problem: Desire to have a fixed scope, fixed date

    solution • Solution: • Never compromise on quality • So what has to give?
  28. F I X E D D AT E • Date

    was arbitrary - based on vendor contract period, not business need.
  29. F I X E D S C O P E

    • Estimation Failure • WBS does not work for projects with high variability. • https:// www.youtube.com/watch? feature=player_detailpag e&v=GiPe1OiKQuk#t=6 • > 2 weeks does not work
  30. F E AT U R E S P L I

    T T I N G “We need feature X” X
  31. “Which of these smaller features do we need?” X1 X2

    X3 X4 F E AT U R E S P L I T T I N G
  32. “We really only need feature X1 and X4” X1 X4

    F E AT U R E S P L I T T I N G
  33. C H A N G I N G T E

    A M D Y N A M I C S P E O P L E P R O B L E M # 2
  34. S U P P LY T H E B E

    S T T O O L S AVA I L A B L E
  35. • Involve them in decisions about WHAT needs to be

    built • Ensure they know WHY it needs to be built • Empower them to decide HOW to build it.
  36. S I L O E D S K I L

    L S E T S P E O P L E P R O B L E M # 3
  37. • Pairing different roles • ATDD - Three Amigos again

    • Ran into problems with organizations outside sphere of influence
  38. S H A R E D O P E R

    AT I O N S S TA F F P E O P L E P R O B L E M # 4
  39. • Attempted to alleviate work by cross-training others so backups

    have sufficient capability • Root cause of this is the organization’s focus on maximizing utilization
  40. Time In Process 0 75 150 225 300 Utilization 10%

    20% 30% 40% 50% 60% 70% 80% 90% 100% :-| :-) :-D :-(
  41. A G I L E & T H E V

    E RY L A R G E O R G A N I Z AT I O N : N A YA N H A J R A T WA L A - N A YA N @ C H I K L I . C O M T W I T T E R : @ N H A J R A T W K C P C S P C S D C S M C J P A P E R F E C T M A T C H