Containers: Practices, Problems, and Possiblilities

Slide 1

Slide 1 text

  Jupyter Team Carol Willing WOSSS19 April 25, 2019 @WillingCarol Containers Practices, Problems, and Possibilities

Slide 2

Slide 2 text

Jupyter Steering Council Core Developer JupyterHub, BinderHub, mybinder.org Python Steering Council Core Developer PSF Fellow Carol Willing

Slide 3

Slide 3 text

Jupyter ‣ Growth ‣ ACM Award ‣ Industry adoption ‣ Creative uses ‣ Open Source Book

Slide 4

Slide 4 text

Project Jupyter Containers 2014-Present @WillingCarol

Slide 5

Slide 5 text

Evolution • jupyter/docker-stacks: Docker containers for common user environments • JupyterHub: Docker containers for ease of deployment and to provide isolated environments for users • Zero to JupyterHub with Kubernetes: Use Kubernetes to scale JupyterHub from a single class to a large scale MOOCs • BinderHub and mybinder.org: Kubernetes and Docker containers used to offer a service @WillingCarol • Environment • Deployment • Scaling • Service

Slide 6

Slide 6 text

zero-to-jupyterhub.readthedocs.io

Slide 7

Slide 7 text

https://blog.jupyter.org/binder-2-0-a-tech-guide-2017-fd40515a3a84 https://elifesciences.org/labs/8653a61d/introducing-binder-2-0-share-your- interactive-research-environment https://www.nature.com/articles/d41586-018-01322-9 mybinder.org

Slide 8

Slide 8 text

Practices @WillingCarol

Slide 9

Slide 9 text

Understand the use case before selecting a container solution. @WillingCarol Practices

Slide 10

Slide 10 text

Converge on solutions for deployment and share openly. @WillingCarol Practices

Slide 11

Slide 11 text

https://zonca.github.io/ @WillingCarol Practices

Slide 12

Slide 12 text

Use standard tools. Don't build software from scratch when standard tools are available. @WillingCarol Practices

Slide 13

Slide 13 text

When running a service, apply Site Reliability best practices. Document processes and perform blameless post- mortems. @WillingCarol Practices

Slide 14

Slide 14 text

Problems @WillingCarol

Slide 15

Slide 15 text

Learning Curve and Software Skills Investment @WillingCarol Problems

Slide 16

Slide 16 text

Data and Access Controls @WillingCarol Problems

Slide 17

Slide 17 text

Scaling @WillingCarol Problems

Slide 18

Slide 18 text

Vendor Lock-in @WillingCarol Problems

Slide 19

Slide 19 text

Possibilities @WillingCarol

Slide 20

Slide 20 text

Tailored Workflows @WillingCarol Possibilities

Slide 21

Slide 21 text

http://genepattern-notebook.org/ @WillingCarol

Slide 22

Slide 22 text

nteract Enterprise data workﬂows

Slide 23

Slide 23 text

Accessible Science @WillingCarol Possibilities

Slide 24

Slide 24 text

Introduction to Mechanical Vibration (Jason Moore & Kenneth Lyons) http://www.moorepants.info/blog/introducing-resonance.html

Slide 25

Slide 25 text

https://jupyter4edu.github.io/jupyter-edu-book/ @WillingCarol

Slide 26

Slide 26 text

Visualization and WebAssembly @WillingCarol Possibilities

Slide 27

Slide 27 text

https://github.com/data- exp-lab/rust-yt-tools/ npm package @data- exp-lab/yt-tools Irber Junior LC. Oxidizing Python: writing extensions in Rust [version 1; not peer reviewed]. F1000Rese arch 2018, 7(ISCB Comm J):955 (poster) (https://doi.org/ 10.7490/ f1000research. 1115726.1) https://github.com/ munkm/widgyts

Slide 28

Slide 28 text

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

Slide 29

Slide 29 text

No content

Slide 30

Slide 30 text

No content

Slide 31

Slide 31 text

Matplotlib documentation Carol Willing papers and photos Project Jupyter nteract project Python website Fernando Perez website ipyvolume project yt project and Madickin Munk Jason Moore

Slide 32

Slide 32 text

  Jupyter Team Carol Willing WOSSS19 April 25, 2019 @WillingCarol Containers Practices, Problems, and Possibilities