Start Small and Scale: Big Data and Jupyter's Ecosystem

Keynote Presentation at PyData LA conference held at Cal State Los Angeles.

Jupyter notebooks have become the de-facto standard as a scientific and data science tool for producing computational narratives. Over five million Jupyter notebooks exist on GitHub today. Beyond the classic Jupyter notebook, Project Jupyter's tools have evolved to provide end to end workflows for research that enable scientists to prototype, collaborate, and scale with ease. JupyterLab, a web-based, extensible, next generation interactive development environment enables researchers to combine Jupyter notebooks, code and data to form computational narratives. JupyterHub brings the power of notebooks to groups of users. It gives users access to computational environments and resources without burdening the users with installation and maintenance tasks. Binder builds upon JupyterHub and provides free, sharable, interactive computing environments to people all around the world.

Transcript

1. @WillingCarol
   Start Small and Scale
   Carol Willing
   PyData LA
   December 5, 2019
   https://speakerdeck.com/willingc
   Big Data and Jupyter's Ecosystem
   

   View Slide

2. @WillingCarol
   Hi! I'm Carol.
   • Python Steering Council
   • Jupyter Steering Council
   • Core Developer, Python, Jupyter,
   nteract
   • PSF Fellow and Former Director
   • Frank Willison Award 2019
   • Open Source Directions Podcast Co-
   host
   2
   

   View Slide

3. @WillingCarol 3
   Core maintainer
   Papermill, Scrapbook, Bookstore, Commuter
   Steering Council, Core Developer
   JupyterHub, BinderHub, mybinder.org
   I love creating tools which
   educate and empower
   people.
   

   View Slide

4. @WillingCarol
   What is
   Data Science
   4
   Start Here
   

   View Slide

5. @WillingCarol 5
   Using data responsibly to
   solve real world issues and
   improve human lives
   

   View Slide

6. @WillingCarol
   Predictions
   at Scale
   6
   A real world tale
   

   View Slide

7. @WillingCarol 7
   San Diego, CA
   

   View Slide

8. @WillingCarol 8
   Tokyo
   

   View Slide

9. @WillingCarol 9
   Sunday Oct 6
   Source: ECMWF
   

   View Slide

10. Copyright: 2019 European
    Union, contains modified
    Copernicus Sentinel data
    2019, processed by
    EUMETSAT
    Super Typhoon
    Hagibis
    View of Super Typhoon
    Hagibis south-west of
    Japan, as captured by
    the Copernicus
    Sentinel-3 satellite on
    08 October at 00:16
    UTC.
    

    View Slide

11. Title Typhoon Hagibis
    Released 10/10/2019 4:45 pm
    Copyright contains modified Copernicus
    Sentinel data (2019), processed by ESA,
    CC BY-SA 3.0 IGO
    

    View Slide

12. Source:Twitter
    

    View Slide

13. @WillingCarol 13
    

    View Slide

14. @WillingCarol 14
    

    View Slide

15. A sign is partially submerged as the Tama River floods during Typhoon Hagibis.
    Source:Getty Images
    Source:Japan Times
    

    View Slide

16. @WillingCarol
    Preparation
    Evacuation
    Safety
    16
    

    View Slide

17. @WillingCarol
    With scale comes
    responsibility
    17
    Reality Check
    

    View Slide

18. @WillingCarol 18
    "Diabetes awareness: character design" by Carlos Hernandez is licensed under CC BY-NC-ND 4.0
    

    View Slide

19. @WillingCarol
    –Kevin Sayer, DexCom CEO
    This whole integration of health care
    data is really going to be the next
    frontier.
    19
    https://www.cnbc.com/2019/11/13/big-data-is-the-next-frontier-for-medicine-says-dexcom-ceo.html
    https://www.businesswire.com/news/home/20191106005764/en/Dexcom-Reports-Quarter-2019-Financial-Results
    

    View Slide

20. @WillingCarol
    Outage
    Midnight Friday: mysterious outage
    Dexcom did not announce there was an
    outage until about 8 a.m. Pacific
    time Saturday, which is 11 a.m. on
    the East Coast, when it posted a
    brief notice on its Facebook page.
    Monday morning: Dexcom Follow partly
    restored
    20
    https://www.nytimes.com/2019/12/02/well/live/Dexcom-G6-diabetes-monitor-outage.html
    Source: https://www.dexcom.com/
    

    View Slide

21. @WillingCarol
    Getting from
    Start to Scale
    21
    The Challenge
    

    View Slide

22. @WillingCarol 22
    jupyter.org
    

    View Slide

23. @WillingCarol 23
    2014
    Now,
    5 years
    later...
    23
    

    View Slide

24. 

    Millions of
    Notebooks
    https://github.com/trending/jupyter-notebook
    Over 5 million
    on GitHub
    

    View Slide

25. @WillingCarol 25
    ‣ Growth
    ‣ ACM Award
    ‣ Industry adoption
    ‣ Creative uses
    ‣ Open Source Book
    https://www.youtube.com/watch?v=qbtDVdEr8SY
    

    View Slide

26. jupyter.org
    

    View Slide

27. @WillingCarol 27
    Start Small
    Deploy and Sustain
    Explore Paths to Scale
    The Roadmap
    

    View Slide

28. @WillingCarol
    Start Small
    28
    Step 1
    

    View Slide

29. @WillingCarol 29
    Source: xkcd
    

    View Slide

30. @WillingCarol 30
    small
    

    View Slide

31. @WillingCarol 31
    Binder 2.0 blog post
    elifesciences: Share
    your interactive
    research environment
    Nature article about
    Binder
    31
    mybinder.org
    Try it. No install needed.
    

    View Slide

32. @WillingCarol 32
    Scale in Production
    

    View Slide

33. @WillingCarol
    Choose your Tools
    33
    

    View Slide

34. @WillingCarol
    JupyterLab
    34
    

    View Slide

35. 35
    jupyter.org demo
    

    View Slide

36. 36
    jupyter.org demo
    

    View Slide

37. @WillingCarol 37
    ReactJS front end
    nteract
    nteract.io
    

    View Slide

38. @WillingCarol
    VS Code
    38
    PyCharm
    

    View Slide

39. @WillingCarol
    Avoid reinventing the wheel
    39
    

    View Slide

40. @WillingCarol
    ecosystem
    40
    

    View Slide

41. @WillingCarol
    Install Promising Libraries
    41
    Use Anaconda
    Use pip
    Use miniconda, conda, and conda-forge
    

    View Slide

42. @WillingCarol
    Start
    42
    Try it in the browser
    Install Libraries
    Choose your tools
    Avoid reinventing the wheel
    Step 1
    

    View Slide

43. @WillingCarol
    Explore Paths to Scale
    43
    Step 2
    

    View Slide

44. @WillingCarol
    Turn no way into
    it's possible
    44
    

    View Slide

45. @WillingCarol
    Community
    45
    Conferences
    Meetups
    PyLadies
    Carpentries
    Photo: Python Sul
    

    View Slide

46. @WillingCarol 46
    Ten Simple Rules
    for Reproducible
    Research in
    Jupyter Notebooks
    Adam Rule et al.
    https://github.com/jupyter-guide/
    ten-rules-jupyter
    https://github.com/jupyter-guide/
    jupyter-guide
    

    View Slide

47. @WillingCarol 47
    build, try, change, repeat
    

    View Slide

48. @WillingCarol 48
    ipyvolume
    https://towardsdatascience.com/multivolume-
    rendering-in-jupyter-with-ipyvolume-cross-
    language-3d-visualization-64389047634a
    

    View Slide

49. @WillingCarol 49
    

    View Slide

50. @WillingCarol 50
    github.com/napari/napari
    napari
    https://ilovesymposia.com/2019/10/24/introducing-napari-
    a-fast-n-dimensional-image-viewer-in-python/
    

    View Slide

51. A pictorial representation of the different tools constituting BinderHub. This image was created by Scriberia for The Turing Way
    community and is used under a CC-BY licence. Zenodo record.
    https://blog.jupyter.org/diving-into-
    leadership-to-build-push-button-code-
    df2a075c9914
    

    View Slide

52. @WillingCarol
    What's new
    52
    Talk Python to Me
    Tracking Jupyter Newsletter
    https://tinyletter.com/TrackingJupyter/archive
    Open Source Directions
    GitHub Trending
    Follow projects on Social Media
    

    View Slide

53. @WillingCarol
    Explore
    53
    Use the ecosystem to learn
    Best practices
    Infrastructure/Analysis
    What's new
    Step 2
    

    View Slide

54. @WillingCarol
    Deploy and Sustain
    54
    Step 3
    

    View Slide

55. @WillingCarol 55
    "Digital World" by NBroekzitter86 is licensed under CC BY 2.0
    

    View Slide

56. @WillingCarol
    Notebooks to web
    56
    https://blog.jupyter.org/and-
    voil%C3%A0-f6a2c08a4a93
    

    View Slide

57. @WillingCarol 57
    nteract
    Papermill
    Scrapbook
    Bookstore
    Commuter
    Production data at scale
    57
    https://medium.com/netflix-techblog/notebook-innovation-591ee3221233
    Bookstore
    

    View Slide

58. @WillingCarol 58
    Papermill - parameterize / run
    Scrapbook - recording / reading
    Bookstore - store notebooks
    Commuter - share notebooks
    Production data at scale
    58
    

    View Slide

59. @WillingCarol 59
    Enterprise data workflows
    59
    

    View Slide

60. zero-to-jupyterhub.readthedocs.io
    

    View Slide

61. @WillingCarol 61 61
    Deploy
    your own
    BinderHub
    

    View Slide

62. 62
    Juliette Taka
    

    View Slide

63. @WillingCarol 63
    From a
    phone in
    the park!
    63
    

    View Slide

64. @WillingCarol
    Pangeo
    64
    https://pangeo.io
    

    View Slide

65. @WillingCarol 65
    

    View Slide

66. @WillingCarol
    Deploy and Sustain
    66
    Workflows
    Document
    Monitor
    Involvement
    Step 3
    

    View Slide

67. @WillingCarol 67
    From Small to Scale
    Keys for Success
    

    View Slide

68. @WillingCarol
    Choose to Start
    68
    

    View Slide

69. @WillingCarol
    Why > how
    69
    

    View Slide

70. @WillingCarol
    Automate the
    Boring Stuff
    70
    

    View Slide

71. @WillingCarol
    Plan
    Execute
    Change
    71
    https://jupyterhub-team-compass.readthedocs.io
    https://github.com/jupyterhub/team-compass
    

    View Slide

72. @WillingCarol
    Consider complexity
    and observability
    72
    

    View Slide

73. @WillingCarol
    People =
    Responsibility
    73
    

    View Slide

74. @WillingCarol 74
    Call to Action
    

    View Slide

75. @WillingCarol
    ecosystem
    75
    

    View Slide

76. @WillingCarol 76
    Using data responsibly to
    solve real world issues and
    improve human lives
    

    View Slide

77. @WillingCarol 77
    Justine Dupont surfs the greatest wave of her life in Nazaré, Portugal
    © Rafael G. Riancho / Red Bull Content Pool
    

    View Slide

78. @WillingCarol 78
    Thank you
    https://speakerdeck.com/willingc
    @WillingCarol
    

    View Slide

79. @WillingCarol 79
    Questions
    https://speakerdeck.com/willingc
    

    View Slide

80. @WillingCarol 80
    Thank you PyData LA
    Project Jupyter Team
    Core Python Team
    PSF
    NumFOCUS
    

    View Slide

81. @WillingCarol
    Attributions
    81
    Attributions on slides.
    Photos
    [7-8] Carol Willing and Linnea Willing
    [14] The Carpentries, Tracy Teal, Bérénice Batut
    [14] Godzilla
    By Toho Company Ltd. (東宝株式会社, Tōhō Kabushiki-kaisha) © 1954 - movie
    poster made by Toho Company Ltd. (東宝株式会社, Tōhō Kabushiki-kaisha),
    Public Domain, https://commons.wikimedia.org/w/index.php?curid=3648684
    

    View Slide