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

How to scale if nothing is for free

How to scale if nothing is for free

André Neubauer

November 28, 2016
Tweet

More Decks by André Neubauer

Other Decks in Technology

Transcript

  1. HOW TO SCALE IF NOTHING
    IS FOR FREE

    View full-size slide

  2. ANDRÉ NEUBAUER - CTO
    2
    „Technology enthusiast, org
    development fanatic and
    strong advocate on modern
    leadership. Motivated by 'Y'!
    Love to work as CTO for
    @Misterspex“
    @devpg

    View full-size slide

  3. EVOLUTIONARY THEORY FOR COMPANIES
    How to scale if nothing is for free

    View full-size slide

  4. EVOLUTIONARY THEORY FOR COMPANIES
    How to scale if nothing is for free
    „Pioneers, Settlers, Town Planner“ by Simon Wardley
    Link: http://blog.gardeviance.org/2015/03/on-pioneers-settlers-town-planners-and.html

    View full-size slide

  5. SCALABILITY IS THE FOUNDATION OF GROWING A
    BUSINESS
    How to scale if nothing is for free
    Economies of scale
    „Scalability is the capability of a system, network, or process to handle a growing amount of
    work, or its potential to be enlarged in order to accommodate that growth.“
    https://en.wikipedia.org/wiki/Scalability

    View full-size slide

  6. SCALING IN THE
    „INDUSTRIAL AGE“

    View full-size slide

  7. ASSEMBLY-LINE WORK

    View full-size slide

  8. „The Principles of Scientific
    Managment“
    Mister Spex Company Profile, Status Mai 2016
    FREDERICK WINSLOW TAYLOR

    View full-size slide

  9. •  Repetitive work
    •  Separation of thought from action
    •  Standardisation of tasks
    •  Scaling by repeating without thinking
    Mister Spex Company Profile, Status Mai 2016
    ASSEMBLY-LINE WORK

    View full-size slide

  10. STEAM ENGINE

    View full-size slide

  11. STEAM ENGINE
    •  One of the most important
    technologies of the Industrial
    Revolution
    •  Enabled faster production
    •  Replaced staff from heavy physical
    work

    View full-size slide

  12. Mister Spex Company Profile, Status Mai 2016
    ECONOMIES OF SCALE

    View full-size slide

  13. LEARNINGS
    •  Scalability (with proportional/ positive effects) is an exception
    •  Most systems don’t scale (in a meaningful degree)
    •  Optimize (till you drop) to scale constant
    •  Change the way of doing things on a different level to achieve
    disproportionately effects

    View full-size slide

  14. WELCOME TO THE „INFORMATION AGE“

    View full-size slide

  15. •  Discontinuous change
    •  Short product life cycles
    •  Continuous price declines
    •  Little customer loyalty
    •  Based on knowledge work
    INFORMATION AGE

    View full-size slide

  16. KNOWLEDGE WORK IS DIFFERENT FROM ASSEMBLY-LINE
    WORK AND DOESN‘T SCALE CONSTANT
    •  Requires creativity
    •  Relies heavily on collaboration between individuals
    •  Doesn’t repeat itself
    •  Is unpredictable but emerges over time

    View full-size slide

  17. SCALING THE TRADITIONAL WAY
    AKA. „ADDING MORE CAPACITY“

    View full-size slide

  18. COLLABORATION IS HARD TO SCALE

    View full-size slide

  19. ECONOMIES OF SCALE

    View full-size slide

  20. HOW TO SCALE
    ANYWAY?!!

    View full-size slide

  21. SCALING THE OTHER WAY
    Economic principle:
    •  Minimum: Variable input, fixed output (default option ... unfortunately)
    •  Maximum: Fixed input, variable output
    Do not ask how much resources you need to make X, but what you can
    achieve with existing resources.
    Scaling does not necessarily mean adding more resources. You can also just use
    the existing resources more efficiently.

    View full-size slide

  22. 10 BEST PRACTICES TO
    SCALE BEFORE ADDING
    MORE RESOURCES

    View full-size slide

  23. #1: FOCUS, FOCUS, FOCUS!

    View full-size slide

  24. •  Strive for throughput rather than
    (local) utilization (Culture of done)
    •  Don‘t worry about things that don‘t
    belong to your value proposition
    •  Use external services/ support/…
    •  Decide for (de-facto) standards
    •  Focus on what moves the needle/
    makes the difference
    #1: FOCUS, FOCUS, FOCUS!

    View full-size slide

  25. #2: CUT PROJECTS!

    View full-size slide

  26. •  Most projects are “black hole
    projects”
    •  Overloaded with features
    •  Consider maintenance!
    #2: CUT PROJECTS!

    View full-size slide

  27. #3: ESTABLISH BUSINESS REVIEWS!

    View full-size slide

  28. •  Compare expectations with actual
    impact
    •  Understand deeply what drives
    success an what causes costs
    •  Derive lessons learned and apply
    them to the next projects
    #3: ESTABLISH BUSINESS REVIEWS!

    View full-size slide

  29. #4: (N)ONE PROCESS TO RULE THEM ALL!

    View full-size slide

  30. •  Conway’s law for processes —> You
    will run projects that match the process
    but not necessarily your needs.
    •  Don’t underestimate costs of process!
    •  Different processes for different
    project
    •  Focus on impact rather than
    utilisation
    #4: (N)ONE PROCESS TO RULE THEM ALL!

    View full-size slide

  31. #5: ORGANIZE FOR COMPLEXITY!

    View full-size slide

  32. “You already have the right people. You
    have a bad organisation”
    Niels Pflaeging
    •  Take cross-functional, self-organized
    and market-driven teams into account
    #5: ORGANIZE FOR COMPLEXITY!
    http://www.slideshare.net/npflaeging/organize-for-complexity-
    keynote-by-niels-pflaeging-at-spark-the-change-torontoca/15
    http://www.slideshare.net/npflaeging/organize-for-complexity-
    keynote-by-niels-pflaeging-at-spark-the-change-torontoca/16

    View full-size slide

  33. “You already have the right people. You
    have a bad organisation”
    Niels Pflaeging
    •  Take cross-functional, self-organized
    and market-driven teams into account
    #5: ORGANIZE FOR COMPLEXITY!
    http://www.organizeforcomplexity.com

    View full-size slide

  34. #6: DON’T MIX UP INNOVATION WITH DAILY WORK!

    View full-size slide

  35. •  Distingush between cash cow and the
    rising star of tomorrow
    •  Have separate standards (regarding
    sustainability, etc.)
    #6: DON’T MIX UP INNOVATION WITH DAILY WORK!

    View full-size slide

  36. #7: RETHINK “THE RIGHT TOOL”!

    View full-size slide

  37. •  Strive for small technology stack,
    don’t increase accidentally
    •  Stay with what you can do aka.
    “Schuster bleibt bei deinen Leisten”
    •  Create space for “experiments”
    #7: RETHINK “THE RIGHT TOOL”!

    View full-size slide

  38. #8: KEEP IT STUPID SIMPLE!
    “Complexity is your enemy. Any fool
    can make something complicated. It is
    hard to keep things simple.”
    Sir Richard Branson

    View full-size slide

  39. •  … not only on software level but
    solution level
    •  Business is complex enough
    •  “Indeed, the ratio of time spent
    reading versus writing is well over 10
    to 1. We are constantly reading old
    code as part of the effort to write new
    code. [… Therefore,] making it easy
    to read makes it easier to write.”
    Robert C. Martin
    #8: KEEP IT STUPID SIMPLE!

    View full-size slide

  40. #9: KNOW YOUR COST DRIVER!

    View full-size slide

  41. •  Important for capacity planning
    •  Distingush between fixed, variable
    and stepped-fixed costs
    •  Stepped-fixed: Fixed over a certain
    range; at particular point, step increase
    in fixed cost
    #9: KNOW YOUR COST DRIVER!

    View full-size slide

  42. #10: CONSIDER MAINTENANCE & LIVE THE LIFECYCLE!
    https://www.researchgate.net/figure/236189469_fig1_Development-of-Software-maintenance-costs-as-percentage-of-total-cost-Floris-and-Harald

    View full-size slide

  43. •  Quality and automation are not
    negotiable (in core business)
    •  Effort for maintenance will eat up
    capacity
    •  Maintenance shouldn’t be the reason for
    scaling
    •  Plan removal of features like new
    projects
    #10: CONSIDER MAINTENANCE & LIVE THE LIFECYCLE!

    View full-size slide

  44. LONG STORY SHORT ...
    •  Just adding more resources doesn‘t work in a knowledge work environment ...
    At least not in a proper way
    •  Give maximum principle at try and rethink way of working continuously
    •  Optimize till you drop before considering additional resources

    View full-size slide

  45. THANKS!
    ANY QUESTIONS?
    How to scale if nothing is for free

    View full-size slide