2019独角兽企业重金招聘Python工程师标准>>> 
JupyterHub on Kubernetes--部署指南
- https://zero-to-jupyterhub.readthedocs.io/en/latest/
JupyterHub is a tool that allows you to quickly utilize cloud computing infrastructure to manage a hub that enables your users to interact remotely with a computing environment that you specify. JupyterHub offers a useful way to standardize the computing environment of a group of people (e.g., for a class of students or an analytics team), as well as allowing people to access the hub remotely.
This growing collection of information will help you set up your own JupyterHub instance. It is in an early stage, so the information and tools may change quickly.
If you have tips or deployments that you would like to share, see Resources from the community. If you see anything that is incorrect or have any questions, feel free to reach out at the issues page.
Getting to Step Zero: your Kubernetes cluster
This section describes a Kubernetes cluster and outlines how to complete Step Zero: your Kubernetes cluster for different cloud providers and infrastructure.
Step Zero: your Kubernetes cluster
- Creating a Kubernetes Cluster
- Step Zero: Kubernetes on Google Cloud
- Step Zero: Kubernetes on Microsoft Azure Container Service (AKS)
- Step Zero: Kubernetes on Amazon Web Services (AWS)
- JupyterHub on Red Hat OpenShift
Creating your JupyterHub
This tutorial starts from Step Zero: your Kubernetes cluster and describes the steps needed for you to create a complete initial JupyterHub deployment. This will use the JupyterHub Helm chart which provides sensible defaults for an initial deployment.
To begin, go to Setting up Helm.
Creating your JupyterHub
- Getting started with JupyterHub
- Setting up Helm
- Setting up JupyterHub
- Turning Off JupyterHub and Computational Resources
Customization Guide
JupyterHub can be configured and customized to fit a variety of deployment requirements. If you would like to expand JupyterHub, customize its setup, increase the computational resources available for users, or change authentication services, this guide will walk you through the steps. See the Helm Chart Configuration Reference for a list of frequently used configurable helm chart fields.
Customization Guide
- Extending your JupyterHub setup
- Applying configuration changes
- Customizing the User Environment
- Use an existing Docker image
- Build a custom Docker image with
repo2docker - Use JupyterLab by default
- Set environment variables
- Pre-populating user’s
$HOMEdirectory with files
- User Resources
- Set user memory and CPU guarantees / limits
- Modifying user storage type and size
- Expanding and contracting the size of your cluster
- User storage in JupyterHub
- How can this process break down?
- Configuration
- Turn off per-user persistent storage
- User Management
- Culling user pods
- Admin Users
- Authenticating Users
Administrator Guide
This section provides information on managing and maintaining a staging or production deployment of JupyterHub. It has considerations for managing cloud-based deployments and tips for maintaining your deployment.
Administrator Guide
- The JupyterHub Architecture
- Debugging Kubernetes
- Debugging commands
- Troubleshooting Examples
- Authentication
- Authenticating with OAuth2
- Full Example of Google OAuth2
- Authenticating with LDAP
- Adding a Whitelist
- Speed and Optimization
- Picking a Scheduler Strategy
- Pre-pulling
- Security
- Reporting a security issue
- HTTPS
- Secure access to Helm
- Audit Cloud Metadata server access
- Delete the Kubernetes Dashboard
- Use Role Based Access Control (RBAC)
- Kubernetes API Access
- Kubernetes Network Policies
- Upgrading your JupyterHub Kubernetes deployment
- Major helm-chart upgrades
- Subtopics
- Troubleshooting
- FAQ
- I thought I had deleted my cloud resources, but they still show up. Why?
- How does billing for this work?
- Advanced Topics
- Ingress
- Arbitrary extra code and configuration in
jupyterhub_config.py - Picking a Scheduler Strategy
- Pre-pulling Images for Faster Startup
- Appendix: Projecting deployment costs
- Cost calculators for cloud providers
- Factors influencing costs
- Interactive Cost Estimator (rough estimate)
- Examples
Resources from the community
This section gives the community a space to provide information on setting up, managing, and maintaining JupyterHub.
Important
We recognize that Kubernetes has many deployment options. As a project team with limited resources to provide end user support, we rely on community members to share their collective Kubernetes knowledge and JupyterHub experiences.
Note
Contibuting to Z2JH. If you would like to help improve the Zero to JupyterHub guide, please see the issues page as well as the contributor guide.
We hope that you will use this section to share deployments with on a variety of infrastructure and for different use cases. There is also a community maintained list of users of this Guide and the JupyterHub Helm Chart.
Please submit a pull request to add to this section. Thanks.
Resources from the community
- Community-authored documentation
- Zero to JupyterHub Gallery of Deployments
- Tips and command snippets
Reference
Reference
- Helm Chart Configuration Reference
- Official JupyterHub and Project Jupyter Documentation
- Tools used in a JupyterHub Deployment
- Glossary
Institutional support
This guide and the associated helm chart would not be possible without the amazing institutional support from the following organizations (and the organizations that support them!)
- UC Berkeley Data Science Division
- Berkeley Institute for Data Science
- Cal Poly, San Luis Obispo
- Simula Research Institute