Anaconda Installation – CentOS Cloud VM
-==Disclaimer==-
This is just a custom guide to a INSTANCE on Google Cloud Platform, this could be more technical and advanced guide, if you do not understand or don’t know what you are doing, I kindly request you to take extreme caution. The components used here are billable. SKYDEVOPS/I/WE do not take any responsibility if the guide causes any serious damage or issues, which you have to take full responsibility.
Assuming that you have access to any cloud provider or any cloud instance like AWS, GCP, Rackspace, DigitalOcean, Linode etc. The instance used has the following specification. This is the Guide to install and configure Anaconda Distribution for Data Science Projects. This guide is a part of Devops for Data Science.
-==System Requirements==-
- CPU: 64-bit 2.8 GHz
- RAM: 4 GB (min)
- Storage: 30 GB.
- License: Free use & redistribution under the terms of Anaconda End User License.
- OS: Windows 7,8,10, macOS 10.10+, or, Ubuntu, RedHat, CentOS 6+, and others.
- System architecture: 64-bit x86, 32-bit x86 with Windows or Linux
-==GUIDE==-
- Part-01: Creating a Instance on the Google Cloud Platform
- Part-02: Installing and configuring Anaconda Distribution – Traditional
- Part-03: Updating and Upgrading Anaconda Distribution
- Part-04: Installing and configuring Anaconda in Docker Container
- Part-05: Minimising the size of the Anaconda Docker Image
- Part-06: Pushing the image to Docker Hub
- Part-07: Automating the entire process using GitHub, Docker Hub
-== Part-01: Creating a Instance on the Google Cloud Platform ==-
Step-01: Name the Instance and select the Region and Zone
Step-02: Select the Machine Type
Step-03: Select the Boot Disk and Storage Options
Step-04: Select OS
Step-05: Verify the Instance
Step-06: Add the DNS configuration
Step-07: Login to the Server
Step-08: Install pre-requisites and update cache
Step-09: Verify the OS release
Step-10: Done – Setting up the instance is completed
-==&&==-
-== Part-02: Installing & configuring Anaconda – Traditional ==-
Step-01: Download the Anaconda Python distribution
Step-02: Python 2.7.x will be used in this guide
Step-03: Create the directory structure for installation
Step-04: Install Anaconda Distribution
Step-05: Accept the License
Step-06: Select the Installation Path
Step-07: Add to the PATH variable
Step-08: Do not install VS Code
Step-09: Verify Anaconda installation
–==&&==-
-== Part-03: Updating and Upgrading Anaconda Distribution ==-
Step-01: Updating Conda
Step-02: Update Anaconda
Step-03: Update All Packages
Step-04: Testing Jupyter Notebooks
-==&&==-
-== Part-04: Installing and configuring Anaconda in Docker Container ==-
Step-01: Installing Docker pre-requisites
Step-02: Add the Docker CE Repo
Step-03: Install Docker Community Edition
Step-04: Starting and Enabling Docker
$ sudo systemctl start docker $ sudo systemctl enable docker
Step-05: Verifying docker installation
Step-06: Dockerfile
# Pre-Built Skydevops CentOS image FROM skydevops/skydevops-centos7-ansible LABEL maintainer="[email protected]" # Updating Centos packages and installing Anaconda RUN yum update && yes|yum upgrade && yum install -y wget bzip2 && wget https://repo.anaconda.com/archive/Anaconda2-5.2.0-Linux-x86_64.sh && bash Anaconda2-5.2.0-Linux-x86_64.sh -b && rm Anaconda2-5.2.0-Linux-x86_64.sh # Set path to conda ENV PATH /root/anaconda2/bin:$PATH # Configuring access to Jupyter RUN mkdir /opt/notebooks jupyter notebook --generate-config --allow-root echo "c.NotebookApp.password = u'sha1:277aa1c52ca6:3a6f71c6f36fc191f238c07e5ca08878837e14f3'" >> /root/.jupyter/jupyter_notebook_config.py # Jupyter listens port: 8888 EXPOSE 8888 # Run Jupyter notebook as Docker main process CMD ["jupyter", "notebook", "--allow-root", "--notebook-dir=/opt/notebooks", "--ip='*'", "--port=8888", "--no-browser"]
Step-07: Create an image using the Dockerfile
$ docker build -t ds-skydevops .
Step-08: Building Image
Step-09: Installing Anaconda
Step-10: Creating Firewall Rule on GCP
Step-11: Firewall Details
Step-12: Assigning it to instance
SKYDEVOPS 