[PYTHON] Easy analysis and sharing with re: dash, an open source data visualization tool Part 1-Installation

• Easy analysis sharing with re: dash, an open source data visualization tool Part 1-Installation
• Easy analysis sharing with re: dash, an open source data visualization tool Part 2 ~ Graph
• Easy analysis sharing with re: dash, an open source data visualization tool Part 3 ~ Google Spreadsheets
• Easy analysis sharing with re: dash, an open source data visualization tool Part 4 ~ pivot table

Numerical data accumulates when operating web services, social games, etc. It is very important to analyze the numerical data and utilize it for the next measure.

It is also good to hit the raw data with MySQL etc., put it out as a numerical value, and paste the graph in Excel, but it takes man-hours to do it every time. Ideally, directors, engineers, and anyone can come up with the numbers they want.

There are tools such as Tableau, but they are too expensive to buy. Even if you are alone, if you want to share it with a team, the license fee will be hundreds of thousands.

So open source re: dash. (Although there is a need to hit SQL)

environment

I'm assuming you created the latest version of Amazon Linux on EC2. Amazon Linux AMI 2015.09.1 (HVM), SSD Volume Type - ami-383c1956

re:dash

re: dash works with data sources such as Redshift and MySQL to visualize that data. Since re: dash itself is a web server, you can check the visualized data with a browser. Moreover, re: dash is open source, so you can build it for free.

However, I tried to build it on mac or Amazon Linux of EC2 by myself, but it is quite difficult to install. If you don't know nginx, gunicorn, supervisord, celery, redis, postgresql, etc., it's difficult to debug when you can't build. In the first place, I have a script for Amazon Linux, but I can't install it because the script is already broken.

Ubuntu and Docker are easy, but I wanted to build it on Amazon Linux. So I forked re: dash and fixed the installation script so that it can be installed on EC2 with Ansible.

Repository

• https://github.com/wapa5pow/ansible-redash-amazon-linux
• Forked re: dash used in ansible-redash-amazon-linux
  • https://github.com/wapa5pow/redash

Constitution

re: dash uses Flask, so if you want to fix it, you can use python. nginx is in front, but I feel that it doesn't have to be separate. The main DB of re: dash is PostgreSQL. I don't need only re: dash for MySQL, but I have it installed because I want to use it as a data source.

Installation procedure

You can install it by doing the following. Please install Ansible.

clone git clone https://github.com/wapa5pow/ansible-redash-amazon-linux

Make hosts one of EC2. EC2 keeps ssh and http ports open cd ansible-redash-amazon-linux vi hosts

Set your-private-key.pem as EC2 SSH private key and do the following ansible-playbook site.yml --private-key=~/.ssh/your-private-key.pem -u ec2-user -i hosts

If Ansible fails, try the same command above again and it will succeed.

How to use re: dash

Login

Log in as admin / admin.

Add data source

re: dash connects to various databases, but to add MySQL installed on the same instance, do as follows.

Creating a query

After creating the data source, create a query.

Visualization

From the issued query, you can create a graph etc. from "+ New Visualization" at the bottom of the same screen.

Creating a dashboard

You can create a dashboard by collecting visualized graphs. If you summarize it here, it seems that regular KPI confirmation will be easy.

Summary

• You can easily build re: dash on EC2 Amazon Linux by using Ansible shown at the beginning.
• re: dash allows you to do free visualizations on the web that you could only build for a fee.
• However, I need to issue SQL, but the visualization itself can be done without SQL

Next time, I will introduce what kind of visualization can be done.