/ipythonnb-s3

IPython notebook manager which seamlessly saves and loads to S3

Primary LanguagePython

IPython notebook S3 store

IPython notebook manager which saves to S3.

Works with IPython 1.1.0

By default the notebook server stores the notebook files in a local directory. This plugin seamlessly stores and reads the notebooks files from and to an S3 bucket. It uses the great boto package to communicate with AWS.

It requires credentials to an AWS account.

How to use

  1. Install the package:

    $ git clone https://github.com/davidbrai/ipythonnb-s3.git
    $ cd ipythonnb-s3
    $ [sudo] python setup.py install
    
  2. Create a profile for your notebook server by running:

    ipython profile create nbserver
    
  3. Edit your ipython_notebook_config.py file (should be in ~/.ipython/profile_nbserver):

    c.NotebookApp.notebook_manager_class = 's3nbmanager.S3NotebookManager'
    c.S3NotebookManager.aws_access_key_id = '<put your aws access key id here>'
    c.S3NotebookManager.aws_secret_access_key = '<put your aws secret access key here>'
    c.S3NotebookManager.s3_bucket = '<put the name of the bucket you want to use>'
    c.S3NotebookManager.s3_prefix = '<put the prefix of the key where your notebooks are stored>'

    If aws_access_key_id and aws_secret_access_key are not supplied then boto will look for the AWS keys in environment variables.

    If you store your notebooks in s3://bucket/simulations/notebooks/ then set c.S3NotebookManager.s3_prefix = 'simulations/notebooks/'

  4. Start notebook server:

    ipython notebook --profile=nbserver