[IPython-dev] Options for cloud-based Jupyterhub for Python training?

Wes Turner wes.turner at gmail.com
Thu Apr 23 11:08:05 EDT 2015


CC'ing this here https://github.com/ipython/ipython/wiki/Install:-Docker

```
## IPython Docker Configurations

IPython Docker Hub Repos: **https://registry.hub.docker.com/u/ipython/**

Source: https://github.com/ipython/docker-notebook

A few ways to install IPython with Docker:

    >>> itertools.product(['IPython', 'IPython Notebook'], ['', '+ SciPy
Stack'])


* IPython:

  ``docker run -it ipython/ipython``

  * https://registry.hub.docker.com/u/ipython/ipython
  * **https://github.com/ipython/ipython/blob/master/Dockerfile**

* IPython + SciPy Stack (``ipython/scipystack``):

  ``docker run -it ipython/scipystack``

  * https://registry.hub.docker.com/u/ipython/scipystack/
    *
https://github.com/ipython/docker-notebook/blob/master/scipystack/Dockerfile
      *
https://github.com/ipython/docker-notebook/blob/master/scipystack/build_scipy_stack.sh

* IPython Notebook (``ipython/notebook``):

  ``docker run -d -p 443:8888 -e "PASSWORD=MakeAPassword" ipython/notebook``

  Then access IPython notebook over SSL/TLS at https://localhost

  * https://registry.hub.docker.com/u/ipython/notebook/
    *
https://github.com/ipython/docker-notebook/blob/master/notebook/Dockerfile

* IPython Notebook + SciPy Stack (``ipython/scipyserver``):

  ``docker run -d -p 443:8888 -e "PASSWORD=MakeAPassword"
ipython/scipyserver``

  * https://registry.hub.docker.com/u/ipython/scipyserver/
      *
https://github.com/ipython/docker-notebook/blob/master/scipyserver/Dockerfile

## Links

Jupyter / JupyterHub

* https://github.com/jupyter/jupyterhub/
  * https://github.com/jupyter/jupyterhub/blob/master/Dockerfile
  * https://registry.hub.docker.com/u/jupyter/
    * https://registry.hub.docker.com/u/jupyter/jupyterhub
      * **https://github.com/jupyter/jupyterhub/wiki/Spawners**
```


On Wed, Apr 22, 2015 at 9:21 PM, Wes Turner <wes.turner at gmail.com> wrote:

> These Dockerfiles might be useful:
> * https://registry.hub.docker.com/u/ipython/scipyserver/
> * https://registry.hub.docker.com/repos/jupyter/
>
> On Wed, Apr 22, 2015 at 5:27 PM, Matthias Bussonnier <
> bussonniermatthias at gmail.com> wrote:
>
>> If you plan on using anaconda, I would directly ask continuum for
>> temporary wakari accounts.
>>
>> You can also build a docker image with your requirement, and if users
>> have a github account it is pretty easy (apparently) to configure
>> jupyterhub with OAuthtenticator plugin with github. Then you "just"
>> need the users github handle. We have that privately  hosted on
>> Rackspace OnMetal servers, but I guess Amazon/Azure or alike is not
>> harder.
>>
>> This is also what Oliver Grisel have used for SciKit Learn Tutorials
>> (but with no logins) during conferences.
>>
>> I think also one of the question is how much resources do you plan to
>> use, are you searching for a "Sponsor" do provide the resources (and
>> eventually fidelize users) , or do you actually have fund to  fiance
>> the server(s) for a week.
>>
>> --
>> M
>>
>> On Wed, Apr 22, 2015 at 3:09 PM, Signell, Richard <rsignell at usgs.gov>
>> wrote:
>> > IPython Folks,
>> > I hope this discussion about https://cloud.sagemath.com hasn't
>> > discouraged discussion of other potential solutions.
>> >
>> > Do folks have experience with digitalocean, cloud9 or other cloud
>> > provisioning that could support Jupyterhub with Anaconda?
>> >
>> > -Rich
>> >
>> > On Wed, Apr 22, 2015 at 6:01 PM, Signell, Richard <rsignell at usgs.gov>
>> wrote:
>> >> It's a technical question.  Everything is completely free and there
>> >> are no license restrictions.
>> >>
>> >> On Wed, Apr 22, 2015 at 5:47 PM, William Stein <wstein at gmail.com>
>> wrote:
>> >>> On Wed, Apr 22, 2015 at 2:42 PM, Signell, Richard <rsignell at usgs.gov>
>> wrote:
>> >>>> We would be fine with Python 2 and Jupyter in
>> >>>> https://cloud.sagemath.com would be fine if those packages were
>> >>>> installed for all projects.
>> >>>>
>> >>>> But some of those packages are tough to build, and getting the right
>> >>>> versions that all play nice together is tricky.   That's why we have
>> >>>> our conda channel.   We have put a lot of effort into that.    Do you
>> >>>> have the ability to install from the binstar.org/ioos channel?
>> >>>
>> >>> I don't know.   Is this a legal or technical question?  Is it
>> >>> completely free?  Are there license restrictions?
>> >>>
>> >>> William
>> >>>
>> >>>>
>> >>>> We don't need heavy duty computational facilities.   This is mostly
>> >>>> about accessing and visualizing met/ocean data.
>> >>>>
>> >>>> On Wed, Apr 22, 2015 at 5:20 PM, William Stein <wstein at gmail.com>
>> wrote:
>> >>>>> On Wed, Apr 22, 2015 at 2:14 PM, Signell, Richard <
>> rsignell at usgs.gov> wrote:
>> >>>>>> William,
>> >>>>>>
>> >>>>>> I mean a python environment with these packages installed from our
>> >>>>>> binstar channel (http://conda.binstar.org):
>> >>>>>>
>> >>>>>> iris
>> >>>>>> mpld3
>> >>>>>> pyoos
>> >>>>>> pandas
>> >>>>>> folium
>> >>>>>> rdflib
>> >>>>>> geojson
>> >>>>>> requests
>> >>>>>> ipython-notebook
>> >>>>>> xlrd
>> >>>>>> mplleaflet
>> >>>>>> oceans
>> >>>>>> qrcode
>> >>>>>> utilities
>> >>>>>
>> >>>>> Does Python 2 versus Python 3 matter?
>> >>>>>
>> >>>>> Also, if you've tried Jupyter in https://cloud.sagemath.com, would
>> it
>> >>>>> be sufficient to meet your needs if it had the above packages
>> >>>>> installed for all projects?
>> >>>>>
>> >>>>> Also, how powerful of computing facilities would be required?
>> >>>>>
>> >>>>>>
>> >>>>>> On Wed, Apr 22, 2015 at 5:09 PM, William Stein <wstein at gmail.com>
>> wrote:
>> >>>>>>> On Wed, Apr 22, 2015 at 2:06 PM, Signell, Richard <
>> rsignell at usgs.gov> wrote:
>> >>>>>>>> Jupyterhub folks,
>> >>>>>>>>
>> >>>>>>>> We are looking for solutions for one week of Python training for
>> about
>> >>>>>>>> 50 met-ocean students at the NSF-funded Unidata Program Center in
>> >>>>>>>> Boulder (in June).  And we are thinking Jupyterhub on the cloud.
>> >>>>>>>>
>> >>>>>>>> We know that Software Carpentry favors students to install on
>> their
>> >>>>>>>> own laptops, but we are leaning toward a cloud solution to
>> demonstrate
>> >>>>>>>> the advantages of computing close to large met-ocean datasets.
>> And we
>> >>>>>>>> could also set up a common environment with packages used in the
>> >>>>>>>> met-ocean community (e.g. http://conda.binstar.org/ioos).
>> >>>>>>>>
>> >>>>>>>> We read this article
>> >>>>>>>>
>> https://developer.rackspace.com/blog/deploying-jupyterhub-for-education/
>> >>>>>>>> which is similar to what we would like to set up, but the setup
>> >>>>>>>> sounded rather harrowing.
>> >>>>>>>>
>> >>>>>>>> Are there other solutions that would take less setup, yet allow
>> us a
>> >>>>>>>> multi-user notebook login with cloud resources with a shared
>> custom
>> >>>>>>>> env?
>> >>>>>>>
>> >>>>>>> Could you elaborate on what you mean by "a shared custom
>> environment".
>> >>>>>>> https://cloud.sagemath.com, which I run, is zero setup, and can
>> easily
>> >>>>>>> handle the load of 50 people (we often have 600-700 simultaneous
>> >>>>>>> users).   However, there's no notion of shared custom
>> environment, so
>> >>>>>>> it might not work at all for you.
>> >>>>>>>
>> >>>>>>>  -- William
>> >>>>>>>
>> >>>>>>>>
>> >>>>>>>> Thanks,
>> >>>>>>>> Rich
>> >>>>>>>>
>> >>>>>>>> --
>> >>>>>>>> Dr. Richard P. Signell   (508) 457-2229
>> >>>>>>>> USGS, 384 Woods Hole Rd.
>> >>>>>>>> Woods Hole, MA 02543-1598
>> >>>>>>>> _______________________________________________
>> >>>>>>>> IPython-dev mailing list
>> >>>>>>>> IPython-dev at scipy.org
>> >>>>>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>> >>>>>>>
>> >>>>>>>
>> >>>>>>>
>> >>>>>>> --
>> >>>>>>> William (http://wstein.org)
>> >>>>>>> _______________________________________________
>> >>>>>>> IPython-dev mailing list
>> >>>>>>> IPython-dev at scipy.org
>> >>>>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>> >>>>>>
>> >>>>>>
>> >>>>>>
>> >>>>>> --
>> >>>>>> Dr. Richard P. Signell   (508) 457-2229
>> >>>>>> USGS, 384 Woods Hole Rd.
>> >>>>>> Woods Hole, MA 02543-1598
>> >>>>>> _______________________________________________
>> >>>>>> IPython-dev mailing list
>> >>>>>> IPython-dev at scipy.org
>> >>>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>> >>>>>
>> >>>>>
>> >>>>>
>> >>>>> --
>> >>>>> William (http://wstein.org)
>> >>>>> _______________________________________________
>> >>>>> IPython-dev mailing list
>> >>>>> IPython-dev at scipy.org
>> >>>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>> >>>>
>> >>>>
>> >>>>
>> >>>> --
>> >>>> Dr. Richard P. Signell   (508) 457-2229
>> >>>> USGS, 384 Woods Hole Rd.
>> >>>> Woods Hole, MA 02543-1598
>> >>>> _______________________________________________
>> >>>> IPython-dev mailing list
>> >>>> IPython-dev at scipy.org
>> >>>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>> >>>
>> >>>
>> >>>
>> >>> --
>> >>> William (http://wstein.org)
>> >>> _______________________________________________
>> >>> IPython-dev mailing list
>> >>> IPython-dev at scipy.org
>> >>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>> >>
>> >>
>> >>
>> >> --
>> >> Dr. Richard P. Signell   (508) 457-2229
>> >> USGS, 384 Woods Hole Rd.
>> >> Woods Hole, MA 02543-1598
>> >
>> >
>> >
>> > --
>> > Dr. Richard P. Signell   (508) 457-2229
>> > USGS, 384 Woods Hole Rd.
>> > Woods Hole, MA 02543-1598
>> > _______________________________________________
>> > IPython-dev mailing list
>> > IPython-dev at scipy.org
>> > http://mail.scipy.org/mailman/listinfo/ipython-dev
>> _______________________________________________
>> IPython-dev mailing list
>> IPython-dev at scipy.org
>> http://mail.scipy.org/mailman/listinfo/ipython-dev
>>
>
>
>
> --
> Wes Turner
> https://westurner.org
> https://wrdrd.com/docs/consulting/knowledge-engineering
>



-- 
Wes Turner
https://westurner.org
https://wrdrd.com/docs/consulting/knowledge-engineering
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/ipython-dev/attachments/20150423/cc000504/attachment.html>


More information about the IPython-dev mailing list