Setting up an AWS analytics server and API in minutes

The steps to stand up an AWS server that can be used to host an analytics dashboard and/or a data feed API

Tony Trevisan


In a previous post, I went over the steps to stand up an AWS server that can be used for automated trading using two trading packages I wrote: rameritrade and etrader. This post will step through the process of setting up a server that can be used to host an analytics dashboard or a data feed API. This post will cover:

The biggest time constraint is the time it takes to install the software. I will link to detailed instructions for all of these steps, but will provide some basic steps that a technical user can follow.

Setting up an AWS Server

AWS has incredible documentation across the website and includes how to Launch an Amazon EC2 Instance.

Once you create an AWS account, you can quickly launch an EC2 server from the dashboard. Rstudio does have an AMI that already has RStudio Server already installed, but this will start with a clean install of Ubuntu. From AWS eC2 dashboard:

Your instance is now running. You will now need to connect to your server to install the necessary software.

Installing the required software

I would highly recommend following the order provided here.

Once your instance is running, click into the details. In the top right corner, click the button that says “Connect”. You should now be on the command line of your server as the ubuntu user. Using the following commands, we are going to create a new user ‘rstudio’ with sudo privileges and then install the software needed.

# Create User with home directory
sudo useradd -m rstudio
# Change user password (this will be your rstudio login credentials)
sudo passwd rstudio

# Grant user sudo priveleges
sudo usermod -a -G sudo rstudio

# Swtich to new user to install software
sudo su rstudio

# Create folder to download software
mkdir downloads
cd downloads

Before going to the next step for installation. We want to open some ports to make sure we can get to the software we install. Under Security, click the security group that is displayed. We are going to open port 3838 for all traffic and port 8787 for just your IP address. You can open port 8787, it will just be less secure. Click “Edit Inbound Rules” and add the following rules. At a later date, you can close port 22 to all traffic but you will need to reach the terminal through either RStudio or an SSH tool like Putty.

You are now ready to install the software. We are going to perform the following steps. The links to detailed instructions are provided. The actual installs may take a few minutes.

# Install R 4.0
sudo add-apt-repository 'deb focal-cran40/'
sudo apt-key adv --keyserver --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9
sudo apt update
sudo apt-get upgrade
gpg --keyserver hkp:// --recv-keys E298A3A825C0D65DFD57CBB651716619E084DAB9
gpg -a --export E298A3A825C0D65DFD57CBB651716619E084DAB9 | sudo apt-key add -
sudo apt install r-base r-base-core r-recommended r-base-dev

# Install RStudio Server
# You may want to visit the install link above to make sure you are getting the latest version
# Make sure you download for Ubuntu 20
sudo apt-get install gdebi-core
sudo gdebi rstudio-server-1.4.1106-amd64.deb

# Install Shiny Server (the first command may take several minutes -especially  Rcpp)
sudo su - -c "R -e \"install.packages('shiny', repos='')\""
sudo gdebi shiny-server-

# Install PostresSQL database and create super user with database
sudo apt install postgresql postgresql-contrib
sudo -u postgres createuser --interactive
  Enter name of role to add: rstudio
  Shall the new role be a superuser? (y/n) y
sudo -u postgres createdb rstudio
# we will need to change the password to work with the scripts
ALTER ROLE rstudio WITH PASSWORD 'rstudio';

# Install Docker
sudo apt install apt-transport-https ca-certificates curl gnupg-agent software-properties-common
curl -fsSL | sudo apt-key add -
sudo add-apt-repository "deb [arch=amd64] $(lsb_release -cs) stable"
sudo apt update
sudo apt install docker-ce docker-ce-cli

Hosting a data API

In order to host a data API, you will first need to collect some data to host. We are going to scrape some financial data from I have already written a script to scrape down data for a few indexes and commodities. You will need to copy down my git repo. Make sure to be in your home directory and then clone the repo down. You will also need to install some more R libraries which could take a few minutes. Again, this will go faster with a t3a.small or t3a.medium.

# Install Linux Libraries for dependencies on R packages
sudo apt-get update -y
sudo apt-get install -y libssl-dev
sudo apt-get install -y libcurl4-openssl-dev
sudo apt-get install -y libxml2-dev
sudo apt-get install -y libpq-dev

# Install necessary R packages
sudo su - -c "R -e \"install.packages(c('httr','tidyverse','RPostgres'))\""

# Clone Git Repo
git clone

# Run R script to download data. This will also save data to a folder for loading data into the database.
/usr/bin/Rscript /home/rstudio/AWS_RShiny_wAPI/code/IndexPull.R

# Sweep data into database - will create tables if none exist
mkdir /home/rstudio/AWS_RShiny_wAPI/db_stg
/usr/bin/Rscript /home/rstudio/AWS_RShiny_wAPI/code/postgres_upd.R

# Setup cron to scrape data automatically and sweep to database
crontab -e

# Press i and then copy these two lines at the bottom of the cron file. 
# They will run twice daily to scrape and sweep 
00 12,22 * * * /usr/bin/Rscript /home/rstudio/AWS_RShiny_wAPI/code/IndexPull.R
10 12,22 * * * /usr/bin/Rscript /home/rstudio/AWS_RShiny_wAPI/code/postgres_upd.R

Now that you have the data stored down. We can setup a docker container that can host the API which would feed your custom Shiny App. In theory, this is extra steps, the Shiny App can be configured to read from the database or read in local files, but if you wanted to host the API or Shiny App on different machines, this is one option.

Below, we are going to pull a docker container and then start the container using a custom docker file. Detailed instructions can be found here.

# Pull Docker File
sudo systemctl start docker
sudo docker pull rstudio/plumber

# If you cloned my git repo, the command below should start a custom docker container
sudo docker build -t customdock /home/rstudio/AWS_RShiny_wAPI/plumber_api/
sudo docker run --rm -p 8000:8000 -v `pwd`/AWS_RShiny_wAPI/plumber_api/app:/app customdock /app/api.R

# Test that the API is now working. Enter the link below into a browser with your IP,Dow

# You can run this command to always keep the API Running
sudo docker run -p 8000:8000 -dit --restart=unless-stopped -v `pwd`/AWS_RShiny_wAPI/plumber_api/app:/app customdock /app/api.R

# Confirm API container is still up
sudo docker ps

Hosting a Shiny App

Now that Shiny is working, the API is up, and the database is running with data populated, we will set up a Shiny app that pulls from the database but can be redirected to pull from the API. I have already built the Shiny app so if you run the commands below, everything should work. This is a very simple app with a single chart and some drop downs to choose from.

# Install Plotly
sudo su - -c "R -e \"install.packages(c('plotly'))\""

# Copy the App to the Shiny Server
sudo cp -rf /home/rstudio/AWS_RShiny_wAPI/example/ /srv/shiny-server/

# Visit your app to make sure it is running

# If there are any issues, you can check the logs
cd /var/log/shiny-server
sudo more ENTER LOG FILE

Connecting to the Postgres Database

Assuming port 5432 was opened to all IPs, there are still some changes to the Postgres configuration files in order to connect to the database with a tool like DB Visualizer. First go to the postgresql.conf file. This can be found in /etc/postgresql/12/main (or whichever version is installed instead of 12). Change the listen address to ‘*’ to allow for all connections. This can be done by entering ‘sudo vi postgresql.conf’ and going to CONNECTIONS AND AUTHENTICATION. Uncomment the listen_addresses and change localhost to ‘*’. Save and close the file.

Next go into pg_hba.conf and enter in ‘host all all md5’ under the first line that calls for TYPE DATABADE USER ADDRESS METHOD. Once saved you need to restart the database (service postgresql restart) and you can then connect from an outside source with proper login information.

Wrapping up

Congratulations! You have now stood up an AWS Server, created and connected to a database, web scraped some data, set up an API, and launched a dashboard. This is only scratching the surface of the amazing features that can be explored with any of these capabilities. Hopefully this gave you a foundation to get started!


For attribution, please cite this work as

Trevisan (2022, May 17). ALT Analytics: Setting up an AWS analytics server and API in minutes. Retrieved from

BibTeX citation

  author = {Trevisan, Tony},
  title = {ALT Analytics: Setting up an AWS analytics server and API in minutes},
  url = {},
  year = {2022}