Table of contents

Enable additional analytics development environments

By default, IBM® Cloud Pak for Data includes Jupyter Notebook Server with Python 3.6. You can optionally install other development environments.

Default environment

The default development environment enables you to create and deploy models using Jupyter Notebook Server. If you use Spark APIs in notebooks, the kernels run in Spark engines in Spark local mode. This development environment includes:
  • Jupyter notebook 5.7.0
  • Anaconda3 5.2 with the conda-forge channel
  • Python 3.6
  • Apache Spark 2.3.2

Additional environments

You can install one or more of following development environments on top of Cloud Pak for Data:
  • Jupyter Notebook Server with Python 2.7
  • Jupyter Notebook Server with Python 3.5
  • RStudio Server with R3.4.3
  • Zeppelin Notebook Server 0.7.3 with Anaconda2 4.4

If you need more information about the development environments, you can get a description of the contents from the add-ons catalog.

Installing analytics development environments

To install a development environment:
  1. SSH into the master node (master-1) of your cluster as root:
    ssh root@MASTER_1_IP
  2. Download the following file from IBM Passport Advantage:
    • For Enterprise Edition, download AnalyticsEnv_x86_nnn.bin.
    • For Cloud Native Edition, download AnalyticEnv_CNE_x86_Vnnn.bin.
  3. Change to the directory where the file was downloaded and make the BIN file executable:
    chmod +x AnalyticsEnv_x86_nnn.bin
  4. Run the BIN file:
    ./AnalyticsEnv_x86_nnn.bin

    This downloads the following TAR file to the ibm/modules directory: analytics-environments.tar.

  5. Change to the /modules directory in your installer files partition. For example, /ibm/modules.
  6. Use the mkdir command to create a subdirectory called analytics-environments.
  7. Extract the contents of the TAR file into the /modules/analytics-environments directory that you created in your installation files partition. For example:
    tar -xvf analytics-environments.tar -C /ibm/modules/analytics-environments
  8. Change to the ibm/InstallPackage/components directory.
  9. Run the deploy.sh script to deploy each environment that you want to use:
    Environment Command
    Jupyter Notebook Server with Python 2.7
    ./deploy.sh /ibm/modules/analytics-environments/0150-py27spark202.tar
    Jupyter Notebook Server with Python 3.5
    ./deploy.sh /ibm/modules/analytics-environments/0160-py35spark221.tar
    RStudio Server with R3.4.3
    ./deploy.sh /ibm/modules/analytics-environments/0180-rstudio.tar
    Zeppelin Notebook Server 0.7.3 with Anaconda2 4.4
    ./deploy.sh /ibm/modules/analytics-environments/0140-zeppelin.tar

What to do next