Here you can find the materials for the "Data Engineering 3: Batch Jobs and APIs" course, part of the MSc in Business Analytics at CEU. For the previous editions, see 2017/2018, 2018/2019, 2019/2020, 2020/2021, 2021/2022, 2022/2023, and 2023/2024.
3 x 2 x 100 mins on April 28, May 5 and 12:
- 13:30 - 15:10 session 1
- 15:10 - 15:40 break
- 15:40 - 17:20 session 2
In-person at the Vienna campus (QS B-421).
Please find in the syllabus
folder of this repository.
- You need a laptop with any operating system and stable Internet connection.
- Please make sure that Internet/network firewall rules are not limiting your access to unusual ports (e.g. 22, 8787, 8080, 8000), as we will heavily use these in the class (can be a problem on a company network). CEU WiFi should have the related firewall rules applied for the class.
- Join the Teams channel dedicated to the class at
ba-de4-2024
with thesymvo7f
team code - When joining remotely, it's highly suggested to get a second monitor where you can follow the online stream, and keep your main monitor for your own work. The second monitor could be an external screen attached to your laptop, e.g. a TV, monitor, projector, but if you don't have access to one, you may also use a tablet or phone to dial-in to the Zoom call.
Goal: learn how to run and schedule R jobs in the cloud.
Excerpts from https://daroczig.github.io/talks
- "A Decade of Using R in Production" (Real Data Science USA - R meetup)
- "Getting Things Logged" (RStudio::conf 2020)
- "Analytics databases in a startup environment: beyond MySQL and Spark" (Budapest Data Forum 2018)
-
Use the following sign-in URL to access the class AWS account: https://657609838022.signin.aws.amazon.com/console
-
Secure your access key(s), other credentials and any login information ...
... because a truly wise person learns from the mistakes of others!
"When I woke up the next morning, I had four emails and a missed phone call from Amazon AWS - something about 140 servers running on my AWS account, mining Bitcoin" -- Hoffman said
"Nevertheless, now I know that Bitcoin can be mined with SQL, which is priceless ;-)" -- Uri Shaked
So set up 2FA (go to IAM / Users / username / Security credentials / Assigned MFA device): https://console.aws.amazon.com/iam
PS probably you do not really need to store any access keys, but you may rely on roles (and the Key Management Service, and the Secrets Manager and so on)
-
Let's use the
eu-west-1
Ireland region
Note: we follow the instructions on Windows in the Computer Lab, but please find below how to access the boxes from Mac or Linux as well when working with the instances remotely.
-
Create (or import) an SSH key in AWS (EC2 / Key Pairs): https://eu-west-1.console.aws.amazon.com/ec2/v2/home?region=eu-west-1#KeyPairs:sort=keyName including using the Owner tag!
-
Get an SSH client:
-
Windows -- Download and install PuTTY: https://www.putty.org
-
Mac -- Install PuTTY for Mac using homebrew or macports
sudo brew install putty sudo port install putty
-
Linux -- probably the OpenSSH client is already installed, but to use the same tools on all operating systems, please install and use PuTTY on Linux too, eg on Ubuntu:
sudo apt install putty
-
-
Convert the generated pem key to PuTTY formatNo need to do this anymore, AWS can provide the key as PPK now.-
GUI: https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/putty.html#putty-private-key
-
CLI:
puttygen key.pem -O private -o key.ppk
-
-
Make sure the key is readable only by your Windows/Linux/Mac user, eg
chmod 0400 key.ppk
-
Create an EC2 instance
- Optional: create an Elastic IP for your box
- Go the the Instances overview at https://eu-west-1.console.aws.amazon.com/ec2/v2/home?region=eu-west-1#Instances:sort=instanceId
- Click "Launch Instance"
- Provide a name for your server (e.g.
daroczig-de3-week1
) and some additional tags for resource tracking, including tagging downstream services, such as Instance and Volumes:- Class:
DE3
- Owner:
daroczig
- Class:
- Pick the
Ubuntu Server 24.04 LTS (HVM), SSD Volume Type
AMI - Pick
t3a.small
(2 GiB of RAM should be enough for most tasks) instance type (see more instance types) - Select your AWS key created above and launch
- Pick a unique name for the security group after clicking "Edit" on the "Network settings"
- Click "Launch instance"
- Note and click on the instance id
-
Connect to the box
- Specify the hostname or IP address
- Specify the "Private key file for authentication" in the Connection category's SSH/Auth/Credentials pane
- Set the username to
ubuntu
on the Connection/Data tab - Save the Session profile
- Click the "Open" button
- Accept & cache server's host key
Alternatively, you can connect via a standard SSH client on a Mac or Linux, something like:
chmod 0400 /path/to/your/pem
ssh -i /path/to/your/pem ubuntu@ip-address-of-your-machine
As a last resort, use "EC2 Instance Connect" from the EC2 dashboard by clicking "Connect" in the context menu of the instance (triggered by right click in the table).
-
Look at the docs: https://www.rstudio.com/products/rstudio/download-server
-
First, we will upgrade the system to the most recent version of the already installed packages. Note, check on the concept of a package manager!
Download Ubuntu
apt
package list:sudo apt update
Optionally upgrade the system:
sudo apt upgrade
And optionally also reboot so that kernel upgrades can take effect.
-
Install R
sudo apt install r-base
To avoid manually answering "Yes" to the question to confirm installation, you can specify the
-y
flag:sudo apt install -y r-base
-
Try R
R
For example:
1 + 4 hist(mtcars$hp) # duh, where is the plot?!
Exit:
q()
Look at the files:
ls ls -latr
-
Install RStudio Server
wget https://download2.rstudio.org/server/jammy/amd64/rstudio-server-2024.12.1-563-amd64.deb sudo apt install -y gdebi-core sudo gdebi rstudio-server-2024.12.1-563-amd64.deb
-
Check process and open ports
rstudio-server status sudo rstudio-server status sudo systemctl status rstudio-server sudo ps aux | grep rstudio sudo apt -y install net-tools sudo netstat -tapen | grep LIST sudo netstat -tapen
-
Look at the docs: http://docs.rstudio.com/ide/server-pro/
-
Confirm that the service is up and running and the port is open
ubuntu@ip-172-31-12-150:~$ sudo netstat -tapen | grep LIST tcp 0 0 0.0.0.0:8787 0.0.0.0:* LISTEN 0 49065 23587/rserver tcp 0 0 0.0.0.0:22 0.0.0.0:* LISTEN 0 15671 1305/sshd tcp6 0 0 :::22 :::* LISTEN 0 15673 1305/sshd
-
Try to connect to the host from a browser on port 8787, eg http://foobar.eu-west-1.compute.amazonaws.com:8787
-
Realize it's not working
-
Open up port 8787 in the security group, by selecting your security group and click "Edit inbound rules":
-
Authentication: http://docs.rstudio.com/ide/server-pro/authenticating-users.html
-
Create a new user:
sudo adduser ceu
-
Login & quick demo:
1+2 plot(mtcars) install.packages('fortunes') library(fortunes) fortune() fortune(200) system('whoami')
-
Reload webpage (F5), realize we continue where we left the browser :)
-
Demo the terminal:
$ whoami ceu $ sudo whoami ceu is not in the sudoers file. This incident will be reported.
-
Grant sudo access to the new user by going back to SSH with
root
access:sudo apt install -y mc sudo mc sudo mcedit /etc/sudoers sudo adduser ceu admin man adduser man deluser
Note 1: might need to relogin / restart RStudio / reload R / reload page .. to force a new shell login so that the updated group setting is applied
Note 2: you might want to add NOPASSWD
to the sudoers
file:
```sh
ceu ALL=(ALL) NOPASSWD:ALL
```
Although also note (3) the related security risks.
-
Custom login page: http://docs.rstudio.com/ide/server-pro/authenticating-users.html#customizing-the-sign-in-page
-
Custom port (e.g. 80): http://docs.rstudio.com/ide/server-pro/access-and-security.html#network-port-and-address
echo "www-port=80" | sudo tee -a /etc/rstudio/rserver.conf sudo rstudio-server restart
Note the pretty outdated R version ... so let's update R by using the apt repo managed by the CRAN team:
wget -qO- https://cloud.r-project.org/bin/linux/ubuntu/marutter_pubkey.asc | sudo tee -a /etc/apt/trusted.gpg.d/cran_ubuntu_key.asc
sudo add-apt-repository "deb https://cloud.r-project.org/bin/linux/ubuntu $(lsb_release -cs)-cran40/"
## fetch most recent list of packages and version, then auto-upgrade
sudo apt-get update && sudo apt-get -y upgrade
Quiz: https://forms.office.com/e/SgME7mHPM9 (5 mins)
Recap on what we covered last week:
- 2FA/MFA in AWS
- Creating EC2 nodes
- Connecting to EC2 nodes via SSH/Putty (note the difference between the PPK and PEM key formats)
- Updating security groups
- Installing RStudio Server
- The difference between R console and Shell
- The use of
sudo
and how to grantroot
(system administrator) privileges - Adding new Linux users, setting password, adding to group
- Messing up package updates and making RStudio Server unusable
Note that you do NOT need to do the instructions below marked with the πͺ emoji -- those have been already done for you, and the related steps are only included below for documenting what has been done and demonstrated in the class.
πͺ Instead of starting from scratch, let's create an Amazon Machine Image (AMI) from the EC2 node we used last week, so that we can use that as the basis of all the next steps:
- Find the EC2 node in the EC2 console
- Right click, then "Image and templates" / "Create image"
- Name the AMI and click "Create image"
- It might take a few minutes to finish
Then you can use the newly created de3-week2
AMI to spin up a new instance for you:
- Go the the Instances overview at https://eu-west-1.console.aws.amazon.com/ec2/v2/home?region=eu-west-1#Instances:sort=instanceId
- Click "Launch Instance"
- Provide a name for your server (e.g.
daroczig-de3-week2
) and some additional tags for resource tracking, including tagging downstream services, such as Instance and Volumes:- Class:
DE3
- Owner:
daroczig
- Class:
- Pick the
de3-week2
AMI - Pick
t3a.small
(2 GiB of RAM should be enough for most tasks) instance type (see more instance types) - Select your AWS key created above and launch
- Select the
de3
security group (granting access to ports 22, 8000, 8080, and 8787) - Click "Advanced details" and select
ceudataserver
IAM instance profile - Click "Launch instance"
- Note and click on the instance id
We'll export the list of IAM users from AWS and create a system user for everyone.
-
Attach a newly created IAM EC2 Role (let's call it
ceudataserver
) to the EC2 box and assign 'Read-only IAM access' (IAMReadOnlyAccess
): -
Install AWS CLI tool (note that using the snap package manager as it was removed from the apt repos):
sudo snap install aws-cli --classic
-
List all the IAM users: https://docs.aws.amazon.com/cli/latest/reference/iam/list-users.html
aws iam list-users
-
Install R packages from JSON parsing and logging (in the next steps) from the apt repo instead of CRAN sources as per https://github.com/eddelbuettel/r2u
wget -q -O- https://eddelbuettel.github.io/r2u/assets/dirk_eddelbuettel_key.asc | sudo tee -a /etc/apt/trusted.gpg.d/cranapt_key.asc sudo add-apt-repository "deb [arch=amd64] https://r2u.stat.illinois.edu/ubuntu noble main" sudo apt update sudo apt-get install -y r-cran-jsonlite r-cran-logger r-cran-glue
Note that all dependencies (let it be an R package or system/Ubuntu package) have been automatically resolved and installed.
Don't forget to click on the brush icon to clean up your terminal output if needed.
Optionally enable
bspm
to enable binary package installations via the traditionalinstall.packages
R function. -
Export the list of users from R:
library(jsonlite) users <- fromJSON(system('aws iam list-users', intern = TRUE)) str(users) users[[1]]$UserName
-
Create a new system user on the box (for RStudio Server access) for every IAM user, set password and add to group:
library(logger) library(glue) for (user in users[[1]]$UserName) { ## remove invalid character user <- sub('@.*', '', user) user <- sub('.', '_', user, fixed = TRUE) log_info('Creating {user}') system(glue("sudo adduser --disabled-password --quiet --gecos '' {user}")) log_info('Setting password for {user}') system(glue("echo '{user}:secretpass' | sudo chpasswd")) # note the single quotes + placement of sudo log_info('Adding {user} to sudo group') system(glue('sudo adduser {user} sudo')) }
Note, you may have to temporarily enable passwordless sudo
for this user (if have not done already) :/
ceu ALL=(ALL) NOPASSWD:ALL
Check users:
readLines('/etc/passwd')
-
Install Jenkins from the RStudio/Terminal: https://www.jenkins.io/doc/book/installing/linux/#debianubuntu
sudo apt install -y fontconfig openjdk-17-jre sudo wget -O /usr/share/keyrings/jenkins-keyring.asc \ https://pkg.jenkins.io/debian-stable/jenkins.io-2023.key echo deb [signed-by=/usr/share/keyrings/jenkins-keyring.asc] \ https://pkg.jenkins.io/debian-stable binary/ | sudo tee \ /etc/apt/sources.list.d/jenkins.list > /dev/null sudo apt-get update sudo apt-get install -y jenkins # check which port is open by java (jenkins) sudo ss -tapen | grep java
-
Open up port 8080 in the related security group
-
Access Jenkins from your browser and finish installation
-
Read the initial admin password from RStudio/Terminal via
sudo cat /var/lib/jenkins/secrets/initialAdminPassword
-
Proceed with installing the suggested plugins
-
Create your first user (eg
ceu
)
-
Note that if loading Jenkins after getting a new IP takes a lot of time, it might be due to
not be able to load the theme.css
as trying to search for that on the previous IP (as per
Jenkins URL setting). To overcome this, wait 2 mins for the theme.css
timeout, login, disable
the dark theme: jenkinsci/dark-theme-plugin#458
Update the security backend to use real Unix users for shared access (if users already created):
And allow jenkins
to authenticate UNIX users and restart:
sudo adduser jenkins shadow
sudo systemctl restart jenkins
Then make sure to test new user access in an incognito window to avoid closing yourself out :)
Optionally you can associate a fixed IP address to your box:
- Allocate a new Elastic IP address at https://eu-west-1.console.aws.amazon.com/ec2/v2/home?region=eu-west-1#Addresses:
- Name this resource by assigning a "Name" tag
- Associate this Elastic IP with your stopped box, then start it
Optionally you can associate a subdomain with your node, using the above created Elastic IP address:
-
Go to Route 53: https://console.aws.amazon.com/route53/home
-
Go to Hosted Zones and click on
ceudata.net
-
Create a new Record, where
- fill in the desired
Name
(subdomain), egde3.ceudata.net
- paste the public IP address or hostname of your server in the
Value
field - click
Create
- fill in the desired
-
Now you will be able to access your box using this custon (sub)domain, no need to remember IP addresses.
To avoid using ports like 8787
and 8080
(and get blocked by the firewall installed on the CEU WiFi), let's configure our services to listen on the standard 80 (HTTP) and potentially on the 443 (HTTPS) port as well, and serve RStudio on the /rstudio
, and Jenkins on the /jenkins
path.
For this end, we will use Nginx as a reverse-proxy, so let's install it first:
sudo apt install -y nginx
First, we need to edit the Nginx config to enable websockets for Shiny apps etc in /etc/nginx/nginx.conf
under the http
section:
map $http_upgrade $connection_upgrade {
default upgrade;
'' close;
}
Then we need to edit the main site's configuration at /etc/nginx/sites-enabled/default
to act as a proxy, which also do some transformations, eg rewriting the URL (removing the /rstudio
path) before hitting RStudio Server:
server {
listen 80;
rewrite ^/rstudio$ $scheme://$http_host/rstudio/ permanent;
location /rstudio/ {
rewrite ^/rstudio/(.*)$ /$1 break;
proxy_pass http://localhost:8787;
proxy_redirect http://localhost:8787/ $scheme://$http_host/rstudio/;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $connection_upgrade;
proxy_read_timeout 20d;
}
}
And restart Nginx:
sudo systemctl restart nginx
Find more information at https://support.rstudio.com/hc/en-us/articles/200552326-Running-RStudio-Server-with-a-Proxy.
Let's see if the port is open on the machine:
sudo ss -tapen|grep LIST
Let's see if we can access RStudio Server on the new path:
curl localhost/rstudio
Now let's see from the outside world ... and realize that we need to open up port 80!
Now we need to tweak the config to support Jenkins as well, but the above Nginx rewrite hack will not work (see https://www.jenkins.io/doc/book/system-administration/reverse-proxy-configuration-troubleshooting/ for more details), so we will just make it a standard reverse-proxy, eg:
server {
listen 80;
rewrite ^/rstudio$ $scheme://$http_host/rstudio/ permanent;
location / {
}
location /rstudio/ {
rewrite ^/rstudio/(.*)$ /$1 break;
proxy_pass http://localhost:8787;
proxy_redirect http://localhost:8787/ $scheme://$http_host/rstudio/;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $connection_upgrade;
proxy_read_timeout 20d;
}
location ^~ /jenkins/ {
proxy_pass http://127.0.0.1:8080/jenkins/;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $remote_addr;
proxy_set_header Host $host;
}
}
And we also need to let Jenkins also know about the custom path, so uncomment Environment="JENKINS_PREFIX=/jenkins"
in /lib/systemd/system/jenkins.service
, then reload the Systemd configs and restart Jenkins:
sudo systemctl daemon-reload
sudo systemctl restart jenkins
See more details at the Jenkins reverse proxy guide.
Optionally, replace the default, system-wide index.html
for folks visiting the root domain without either the rstudio
or jenkins
path (note that instead of the editing the file, which might be overwritten with package updates, it would be better to create a new HTML file and refer that from the Nginx configuration, but we will keep it simple and dirty for now):
echo "Welcome to DE3! Are you looking for <code>/rstudio</code> or <code>/jenkins</code>?" | sudo tee /usr/share/nginx/html/index.html
Then restart Jenkins, and good to go!
It might be useful to also proxy port 8000 for future use via updating the Nginx config to:
server {
listen 80;
rewrite ^/rstudio$ $scheme://$http_host/rstudio/ permanent;
location / {
}
location /rstudio/ {
rewrite ^/rstudio/(.*)$ /$1 break;
proxy_pass http://localhost:8787;
proxy_redirect http://localhost:8787/ $scheme://$http_host/rstudio/;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection $connection_upgrade;
proxy_read_timeout 20d;
}
location ^~ /jenkins/ {
proxy_pass http://127.0.0.1:8080/jenkins/;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $remote_addr;
proxy_set_header Host $host;
}
location ^~ /8000/ {
rewrite ^/8000/(.*)$ /$1 break;
proxy_pass http://127.0.0.1:8000;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $remote_addr;
proxy_set_header Host $host;
}
}
This way you can access the above services via the below URLs:
RStudio Server:
Jenkins:
Port 8000:
If you cannot access RStudio Server on port 80, you might need to restart nginx
as per above.
Next, set up SSL either with Nginx or placing an AWS Load Balancer in front of the EC2 node.
Install ggplot2
or your preferred Python packages to replicate the below steps that we will automate later:
-
Install the
devtools
R package and a few others (binary distribution) in the RStudio/Terminal:sudo apt-get install -y r-cran-devtools r-cran-data.table r-cran-httr r-cran-jsonlite r-cran-data.table r-cran-stringi r-cran-stringr r-cran-glue r-cran-logger r-cran-snakecase
-
Switch back to the R console and install the
binancer
R package from GitHub to interact with crypto exchanges (note the extra dependency to be installed from CRAN, no need to update any already installed package):devtools::install_github('daroczig/binancer', upgrade = FALSE)
-
First steps with live data: load the
binancer
package and then use thebinance_klines
function to get the last 3 hours of Bitcoin price changes (in USD) with 1-minute granularity -- resulting in an object like:> str(klines) Classes βdata.tableβ and 'data.frame': 180 obs. of 12 variables: $ open_time : POSIXct, format: "2020-03-08 20:09:00" "2020-03-08 20:10:00" "2020-03-08 20:11:00" "2020-03-08 20:12:00" ... $ open : num 8292 8298 8298 8299 8298 ... $ high : num 8299 8299 8299 8299 8299 ... $ low : num 8292 8297 8297 8298 8296 ... $ close : num 8298 8298 8299 8298 8299 ... $ volume : num 25.65 9.57 20.21 9.65 24.69 ... $ close_time : POSIXct, format: "2020-03-08 20:09:59" "2020-03-08 20:10:59" "2020-03-08 20:11:59" "2020-03-08 20:12:59" ... $ quote_asset_volume : num 212759 79431 167677 80099 204883 ... $ trades : int 371 202 274 186 352 271 374 202 143 306 ... $ taker_buy_base_asset_volume : num 13.43 5.84 11.74 7.12 15.24 ... $ taker_buy_quote_asset_volume: num 111430 48448 97416 59071 126493 ... $ symbol : chr "BTCUSDT" "BTCUSDT" "BTCUSDT" "BTCUSDT" ... - attr(*, ".internal.selfref")=<externalptr>
Click here for the code generating the above ...
library(binancer) klines <- binance_klines('BTCUSDT', interval = '1m', limit = 60*3) str(klines) summary(klines$close)
-
Visualize the data, eg on a simple line chart:
Click here for the code generating the above ...
library(ggplot2) ggplot(klines, aes(close_time, close)) + geom_line()
-
Now create a candle chart, something like:
Click here for the code generating the above ...
library(scales) ggplot(klines, aes(open_time)) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) # RdYlGn
-
Compare prices of 4 currencies (eg BTC, ETH, BNB and XRP) in the past 24 hours on 15 mins intervals:
Click here for the code generating the above ...
library(data.table) klines <- rbindlist(lapply( c('BTCUSDT', 'ETHUSDT', 'BNBUSDT', 'XRPUSDT'), binance_klines, interval = '15m', limit = 4*24)) ggplot(klines, aes(open_time)) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_color_manual(values = c('#1a9850', '#d73027')) + facet_wrap(~symbol, scales = 'free', nrow = 2)
-
Some further useful functions:
binance_ticker_all_prices()
binance_coins_prices()
binance_credentials
andbinance_balances
-
Create an R script that reports and/or plots on some cryptocurrencies, ideas:
- compute the (relative) change in prices of cryptocurrencies in the past 24 / 168 hours
- go back in time 1 / 12 / 24 months and "invest" $1K in BTC and see the value today
- write a bot buying and selling crypto on a virtual exchange
Let's schedule a Jenkins job to check on the Bitcoin prices every hour!
-
Visit Jenkins using the
/jenkins
URL path of your instance's public IP address -
Use your UNIX username and a password to log in
If logging in takes a long time, it might be due to the Dark Theme plugin trying to load a CSS file from an outdated location absed on the Jenkins URL configured at
/jenkins/manage/configure
. To overcome this issue, wait it out at the above URL and update the IP address to your new dynamic IP address, or disable the plugin at/jenkins/manage/pluginManager/installed
and then restart Jenkins at the bottom of the page. -
Create a "New Item" (job):
-
Enter the name of the job:
get current Bitcoin price
-
Pick "Freestyle project"
-
Click "OK"
-
Add a new "Execute shell" build step
-
Enter the below command to look up the most recent BTC price
R -e "library(binancer);binance_coins_prices()[symbol == 'BTC', usd]"
-
Run the job
-
-
Debug & figure out what's the problem ...
-
Install R packages system wide from RStudio/Terminal (more on this later):
Either start R in the terminal as the root user (via
sudo R
) and run the previousdevtools::install_github
command there, or with a one-liner:sudo Rscript -e "library(devtools);withr::with_libpaths(new = '/usr/local/lib/R/site-library', install_github('daroczig/binancer', upgrade = FALSE))"
-
Rerun the job
-
Create an R script with the below content and save on the server, eg as
/home/ceu/bitcoin-price.R
:library(binancer) prices <- binance_coins_prices() paste('The current Bitcoin price is', prices[symbol == 'BTC', usd])
-
Follow the steps from the Schedule R commands section to create a new Jenkins job, but instead of calling
R -e "..."
in shell step, reference the above R script usingRscript
instead:Rscript /home/ceu/de3.R
Alternatively, you could also install little R for this purpose:
sudo apt install -y r-cran-littler r /home/ceu/de3.R
Note the permission error, so let's add the
jenkins
user to theceu
group:sudo adduser jenkins ceu
Then restart Jenkins from the RStudio Server terminal:
sudo systemctl restart jenkins
A better solution will be later to commit our R script into a git repo, and make it part of the job to update from the repo.
-
Create an R script that generates a candlestick chart on the BTC prices from the past hour, saves as
btc.png
in the workspace, and update every 5 minutes!Example solution for the above ...
library(binancer) library(ggplot2) library(scales) klines <- binance_klines('BTCUSDT', interval = '1m', limit = 60) g <- ggplot(klines, aes(open_time)) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) ggsave('btc.png', plot = g, width = 10, height = 5)
-
Enter the name of the job:
Update BTC candlestick chart
-
Pick "Freestyle project"
-
Click "OK"
-
Add a new "Execute shell" build step
-
Enter the below command to look up the most recent BTC price
Rscript /home/ceu/plot.R
-
Run the job
-
Look at the workspace that can be accessed from the sidebar menu of the job.
-
-
Use a git repository to store the R scripts and fetch the most recent version on job start:
- Configure the Jenkins job to use "Git" in the "Source Code Management" section, and use e.g. https://gist.github.com/daroczig/e5d3ee3664549932bb7f23ce8e93e472 as the repo URL, and specify the branch (
main
). - Update the Execute task section to refer to the
btcprice.R
file of the repo instead of the hardcoded local path. - Make edits to the repo, e.g. update lookback to 3 hours and check a future job output.
- Configure the Jenkins job to use "Git" in the "Source Code Management" section, and use e.g. https://gist.github.com/daroczig/e5d3ee3664549932bb7f23ce8e93e472 as the repo URL, and specify the branch (
-
Set up e-mail notifications via eg https://app.mailjet.com/signin
-
πͺ Sign up, confirm your e-mail address and domain
-
πͺ Take a note on the SMTP settings, eg
- SMTP server: in-v3.mailjet.com
- Port: 465
- SSL: Yes
- Username: ***
- Password: ***
-
πͺ Configure Jenkins at http://de3.ceudata.net/jenkins/configure
-
Set up the default FROM e-mail address at "System Admin e-mail address": [email protected]
-
Search for "Extended E-mail Notification" and configure
- SMTP Server
- Click "Advanced"
- Check "Use SMTP Authentication"
- Enter User Name from the above steps
- Enter Password from the above steps
- Check "Use SSL"
- SMTP port: 465
-
-
Set up "Post-build Actions" in Jenkins: Editable Email Notification - read the manual and info popups, configure to get an e-mail on job failures and fixes
-
Configure the job to send the whole e-mail body as the deault body template for all outgoing emails
${BUILD_LOG, maxLines=1000}
-
-
Look at other Jenkins plugins, eg the Slack Notifier: https://plugins.jenkins.io/slack
We need a persistent storage for our Jenkins jobs ... let's give a try to a key-value database:
-
πͺ Install server
sudo apt install redis-server ss -tapen | grep LIST
Test using the CLI tool:
redis-cli get foo redis-cli set foo 42 redis-cli get foo redis-cli del foo
-
πͺ Install an R client
Although we could use the
RcppRedis
available on CRAN, so thus also in our already added apt repo, butrredis
provides some convenient helpers that we plan to use, so we are going to install that as well from a custom R package repository to also demonstrate howdrat
works:sudo apt install -y r-cran-rcppredis sudo Rscript -e "withr::with_libpaths(new = '/usr/local/lib/R/site-library', install.packages('rredis', repos=c('https://ghrr.github.io/drat', 'https://cloud.r-project.org')))"
-
Interact from R
## set up and initialize the connection to the local redis server library(rredis) redisConnect() ## set/get values redisSet('foo', 'bar') redisGet('foo') ## increment and decrease counters redisIncr('counter') redisIncr('counter') redisIncr('counter') redisGet('counter') redisDecr('counter') redisDecr('counter2') ## get multiple values at once redisMGet(c('counter', 'counter2')) ## list all keys redisKeys() ## drop all keys redisDelete(redisKeys())
For more examples and ideas, see the rredis
package vignette or try the interactive, genaral (not R-specific) redis tutorial.
-
Exercises
- Create a Jenkins job running every minute to cache the most recent Bitcoin and Ethereum prices in Redis
- Write an R script in RStudio that can read the Bitcoin and Ethereum prices from the Redis cache
Example solution ...
library(binancer)
library(data.table)
prices <- binance_coins_prices()
library(rredis)
redisConnect()
redisSet('username:price:BTC', prices[symbol == 'BTC', usd])
redisSet('username:price:ETH', prices[symbol == 'ETH', usd])
redisGet('username:price:BTC')
redisGet('username:price:ETH')
redisMGet(c('username:price:BTC', 'username:price:ETH'))
Example solution using a helper function doing some logging ...
library(binancer)
library(logger)
library(rredis)
redisConnect()
store <- function(s) {
## TODO use the checkmate pkg to assert the type of symbol
log_info('Looking up and storing {s}')
value <- binance_coins_prices()[symbol == s, usd]
key <- paste('username', 'price', s, sep = ':')
redisSet(key, value)
log_info('The price of {s} is {value}')
}
store('BTC')
store('ETH')
## list all keys with the "price" prefix and lookup the actual values
redisMGet(redisKeys('username:price:*'))
More on databases at the "Mastering R" class in the Spring semester ;)
-
Join the #bots-bots-bots channel in the DE3 course's MS Teams
-
Click on "Connectors" in the channel's config and add an incoming webhook with your username and optional logo, store the URL for later use
-
πͺ Install the
teamr
package from CRANsudo apt install -y r-cran-teamr
-
Use the webhook URL to send messages to the channel from R:
library(teamr) webhook_url <- "https://ceuedu.webhook.office.com/webhookb2/..." # create new connector card cc <- connector_card$new(hookurl = webhook_url) # set the title and text of the message cc$title("Hi from R!") cc$text("This is a test message sent using the `teamr` package.") # send the message cc$send()
Quiz: https://forms.office.com/e/wRAxGqirdV (5 mins)
Recap on what we covered last week:
- Amazon Machine Image
- Shared RStudio Server
- Settup up Jenkins
- Reverse proxy for more conenvient access to services
- Scheduling R commands in Jenkins
- Scheduling R script in Jenkins
- Further Jenkins features, such as e-mail notifications and git integration
- Introduction to Redis
- Interacting with MS Teams
New server for this week:
- Go the the Instances overview at https://eu-west-1.console.aws.amazon.com/ec2/v2/home?region=eu-west-1#Instances:sort=instanceId
- Click "Launch Instance"
- Provide a name for your server (e.g.
daroczig-de3-week3
) and some additional tags for resource tracking, including tagging downstream services, such as Instance and Volumes:- Class:
DE3
- Owner:
daroczig
- Class:
- Pick the
de3
AMI - Pick
t3a.small
(2 GiB of RAM should be enough for most tasks) instance type (see more instance types) - Select your AWS key created above and launch
- Select the
de3
security group (granting access to ports 22, 8000, 8080, and 8787) - Click "Advanced details" and select
ceudataserver
IAM instance profile - Click "Launch instance"
- Note and click on the instance id
-
Do NOT store the webhook URL in plain-text (e.g. in your R script)!
-
Let's use Amazon's Key Management Service: https://github.com/daroczig/CEU-R-prod/raw/2017-2018/AWR.Kinesis/AWR.Kinesis-talk.pdf (slides 73-75)
-
πͺ Instead of using the Java SDK referenced in the above talk, let's install
boto3
Python module and use viareticulate
:sudo apt install -y python3-boto3 sudo apt install -y r-cran-reticulate r-cran-botor
Let's also let R know that we want to use the globally installed Python interpreter and its packages instead of setting up local virtual environments by adding the following to your
/etc/R/Renviron.site
file:RETICULATE_PYTHON=/usr/bin/python3
-
πͺ Create a key in the Key Management Service (KMS):
alias/de3
-
πͺ Grant access to that KMS key by creating an EC2 IAM role at https://console.aws.amazon.com/iam/home?region=eu-west-1#/roles with the
AWSKeyManagementServicePowerUser
policy and explicit grant access to the key in the KMS console -
πͺ Attach the newly created IAM role if not yet done
-
Use this KMS key to encrypt the Slack token:
library(botor) botor(region = 'eu-west-1') kms_encrypt(webhook_url, key = 'alias/de3')
Note, if R asks you to install Miniconda, say NO, as Python3 and the required packages have already been installed system-wide.
-
Store the ciphertext and use
kms_decrypt
to decrypt later, see egkms_decrypt("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")
-
πͺ Alternatively, use the AWS Parameter Store or Secrets Manager, see eg https://eu-west-1.console.aws.amazon.com/systems-manager/parameters/?region=eu-west-1&tab=Table and granting the
AmazonSSMReadOnlyAccess
policy to your IAM role or user.ssm_get_parameter('/teams/daroczig')
-
Store your own webhook in the Parameter Store and use it in your R script.
- Create a Jenkins job to alert if Bitcoin price is below $80K or higher than $100K.
- Limit this job to alert only once per hour, but alert should kick off as soon as possible (so don't schedule to run hourly, instead use a state to track when the last alert was sent).
- Create a Jenkins job to alert if Bitcoin price changed more than $200 in the past hour.
- Create a Jenkins job to alert if Bitcoin price changed more than 5% in the past day.
- Create a Jenkins job running hourly to generate a candlestick chart on the price of BTC and ETH.
Example solution for the first exercise ...
## get data right from the Binance API
library(binancer)
btc <- binance_klines('BTCUSDT', interval = '1m', limit = 1)$close
## or from the local cache (updated every minute from Jenkins as per above)
library(rredis)
redisConnect()
btc <- redisGet('username:price:BTC')
## log whatever was retreived
library(logger)
log_info('The current price of a Bitcoin is ${btc}')
## get the last alert time
last_alert <- redisGet('username:alert:last')
if (is.null(last_alert)) {
last_alert <- 0
}
since_last_alert <- as.numeric(
difftime(Sys.time(), last_alert,
units = "secs"))
## send alert
if (since_last_alert >= 60 && (btc < 80000 | btc > 100000)) {
library(botor)
botor(region = 'eu-west-1')
webhook_url <- ssm_get_parameter('/teams/username')
library(teamr)
cc <- connector_card$new(hookurl = webhook_url)
cc$title('Bitcoin price alert!')
cc$text(paste('The current price of a Bitcoin is:', btc))
cc$send()
}
-
πͺ Install plumber: rplumber.io
sudo apt install -y r-cran-plumber
-
Create an API endpoint to show the min, max and mean price of a BTC in the past hour!
Create
~/plumber.R
with the below content:library(binancer) #* BTC stats #* @get /btc function() { klines <- binance_klines('BTCUSDT', interval = '1m', limit = 60L) klines[, .(min = min(close), mean = mean(close), max = max(close))] }
Start the plumber application wither via clicking on the "Run API" button or the below commands:
library(plumber) pr("plumber.R") %>% pr_run(host='0.0.0.0', port=8000)
-
Add a new API endpoint to generate the candlestick chart with dynamic symbol (default to BTC), interval and limit! Note that you might need a new
@serializer
, function arguments, and type conversions as well.Example solution for the above ...
library(binancer) library(ggplot2) library(scales) #* Generate plot #* @param symbol coin pair #* @param interval:str enum #* @param limit integer #* @get /klines #* @serializer png function(symbol = 'BTCUSDT', interval = '1m', limit = 60L) { klines <- binance_klines(symbol, interval = interval, limit = as.integer(limit)) # NOTE int conversion library(scales) p <- ggplot(klines, aes(open_time)) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) # RdYlGn print(p) }
-
Add a new API endpoint to generate a HTML report including both the above!
Example solution for the above ...
πͺ Update the
markdown
package:sudo apt install -y r-cran-markdown
Create an R markdown for the reporting:
--- title: "report" output: html_document date: "`r Sys.Date()`" --- ```{r setup, include=FALSE} knitr::opts_chunk$set(echo = FALSE, warning=FALSE) library(binancer) library(ggplot2) library(scales) library(knitr) klines <- function() { binance_klines('BTCUSDT', interval = '1m', limit = 60L) } ``` Bitcoin stats: ```{r stats} kable(klines()[, .(min = min(close), mean = mean(close), max = max(close))]) ``` On a nice plot: ```{r plot} ggplot(klines(), aes(open_time, )) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) ```
And the plumber file:
library(binancer) library(ggplot2) library(scales) library(rmarkdown) library(plumber) #' Gets BTC data from the past hour #' @return data.table klines <- function() { binance_klines('BTCUSDT', interval = '1m', limit = 60L) } #* BTC stats #* @get /stats function() { klines()[, .(min = min(close), mean = mean(close), max = max(close))] } #* Generate plot #* @get /plot #* @serializer png function() { p <- ggplot(klines(), aes(open_time, )) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) # RdYlGn print(p) } #* Generate HTML #* @get /report #* @serializer html function(res) { filename <- tempfile(fileext = '.html') on.exit(unlink(filename)) render('report.Rmd', output_file = filename) include_file(filename, res) }
Run via:
library(plumber) pr('plumber.R') %>% pr_run(port = 8000)
Try to DRY (don't repeat yourself!) this up as much as possible.
Why API? Why R-based API? Examples
- adtech
- healthtech
-
Write an R script that provides 3 API endpoints (look up examples from past week!):
/stats
reports on the min/mean/max BTC price from the past 3 hours/plot
generates a candlestick chart on the price of BTC from past 3 hours/report
generates a HTML report including both the above
Example solution for the above ...
πͺ Update the
markdown
package:sudo apt install -y r-cran-markdown
Create an R markdown for the reporting:
--- title: "report" output: html_document date: "`r Sys.Date()`" --- ```{r setup, include=FALSE} knitr::opts_chunk$set(echo = FALSE, warning=FALSE) library(binancer) library(ggplot2) library(scales) library(knitr) klines <- function() { binance_klines('BTCUSDT', interval = '1m', limit = 60L) } ``` Bitcoin stats: ```{r stats} kable(klines()[, .(min = min(close), mean = mean(close), max = max(close))]) ``` On a nice plot: ```{r plot} ggplot(klines(), aes(open_time, )) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) ```
And the plumber file:
library(binancer) library(ggplot2) library(scales) library(rmarkdown) library(plumber) #' Gets BTC data from the past hour #' @return data.table klines <- function() { binance_klines('BTCUSDT', interval = '1m', limit = 60L) } #* BTC stats #* @get /stats function() { klines()[, .(min = min(close), mean = mean(close), max = max(close))] } #* Generate plot #* @get /plot #* @serializer png function() { p <- ggplot(klines(), aes(open_time, )) + geom_linerange(aes(ymin = open, ymax = close, color = close < open), size = 2) + geom_errorbar(aes(ymin = low, ymax = high), size = 0.25) + theme_bw() + theme('legend.position' = 'none') + xlab('') + ggtitle(paste('Last Updated:', Sys.time())) + scale_y_continuous(labels = dollar) + scale_color_manual(values = c('#1a9850', '#d73027')) # RdYlGn print(p) } #* Generate HTML #* @get /report #* @serializer html function(res) { filename <- tempfile(fileext = '.html') on.exit(unlink(filename)) render('report.Rmd', output_file = filename) include_file(filename, res) }
Run via:
library(plumber) pr('plumber.R') %>% pr_run(port = 8000)
-
Try to DRY (don't repeat yourself!) this up as much as possible.
-
Bundle all the scripts into a single Docker image:
a. πͺ Install Docker:
curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo gpg --dearmor -o /etc/apt/keyrings/docker.gpg echo \ "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.gpg] \ https://download.docker.com/linux/ubuntu \ $(lsb_release -cs) stable" | sudo tee /etc/apt/sources.list.d/docker.list > /dev/null sudo apt-get update sudo apt-get install -y docker-ce
b. Create a new file named
Dockerfile
(File/New file/Text file to avoid auto-adding theR
file extension) with the below content to add the required files and set the default working directory to the same folder:FROM rstudio/plumber RUN apt-get update && apt-get install -y pandoc && apt-get clean && rm -rf /var/lib/apt/lists/ RUN install2.r ggplot2 RUN installGithub.r daroczig/binancer ADD report.Rmd /app/report.Rmd ADD plumber.R /app/plumber.R EXPOSE 8000 WORKDIR /app
Note the step installing the required R packages!
c. Build the Docker image:
sudo docker build -t btc-report-api .
d. Run a container based on the above image:
sudo docker run -p 8000:8000 -ti btc-report-api plumber.R
e. Realize that it's not working, as we need to install
rmarkdown
package, so update the aboveDockerfile
on line 4, rebuild the image and run again.f. Test by visiting the
8000
port or the Nginx proxy at http://de3.ceudata.net/8000, e.g. Swagger docs at http://de3.ceudata.net/8000/__docs__/#/default/get_report or an endpoint directly at http://de3.ceudata.net/8000/report.
Now let's make the above created and tested Docker image available outside of the RStudio Server by uploading the Docker image to Elastic Container Registry (ECR):
-
πͺ Create a new private repository at https://eu-west-1.console.aws.amazon.com/ecr/home?region=eu-west-1, call it
de3-example-api
-
πͺ Assign the
EC2InstanceProfileForImageBuilderECRContainerBuilds
policy to theceudataserver
IAM role so that we get RW access to the ECR repositories. Tighten this role up in prod! -
πͺ Let's login to ECR on the RStudio Server so that we can upload the Docker image:
aws ecr get-login-password --region eu-west-1 | sudo docker login --username AWS --password-stdin 657609838022.dkr.ecr.eu-west-1.amazonaws.com
-
πͺ Tag the already build Docker image for upload:
sudo docker tag btc-report-api:latest 657609838022.dkr.ecr.eu-west-1.amazonaws.com/de3-example-api:latest
-
πͺ Push the Docker image:
sudo docker push 657609838022.dkr.ecr.eu-west-1.amazonaws.com/de3-example-api:latest
-
Check the Docker repository at https://eu-west-1.console.aws.amazon.com/ecr/repositories/private/657609838022/de3-example-api?region=eu-west-1
-
Go to the Elastic Container Service (ECS) dashboard at https://eu-west-1.console.aws.amazon.com/ecs/home?region=eu-west-1#/
-
Create a task definition for the Docker run:
- Task name:
btc-api
- Container name:
api
- Image URI:
657609838022.dkr.ecr.eu-west-1.amazonaws.com/de3-example-api
- Container port: 8000
- Command in the Docker configuration:
plumber.R
- Review Task size, but default values should fine for this simple task
- Task name:
-
Create a new cluster, call it
BTC_API
, using Fargate. Don't forget to add theClass
tag! -
Create a Service in the newly created Cluster at https://eu-west-1.console.aws.amazon.com/ecs/v2/clusters/btc-api/services?region=eu-west-1
- Compute option can be "Launch type" for now
- Specify the Task Family as
btc-api
- Provide the same as service name
- Use the
de3
security group - Create a load balancer listening on port 80 (would need to create an SSL cert for HTTPS), and specify
/stats
as the healthcheck path, with a 10 seconds grace period - Test the deployed service behind the load balancer, e.g. https://btc-api-1417435399.eu-west-1.elb.amazonaws.com/report
The goal of this assignment is to confirm that you have a general understanding on how to build data pipelines using Amazon Web Services and R, and can actually implement a stream processing application (either running in almost real-time or batched/scheduled way) or R-based API in practice.
To minimize the system administration and some of the already-covered engineering tasks for the students, the below pre-configured tools are provided as free options, but students can decide to build their own environment (on the top of or independently from these) and feel free to use any other tools:
de3
Amazon Machine Image that you can use to spin up an EC2 node with RStudio Server, Shiny Server, Jenkins, Redis and Docker installed & pre-configured (use your AWS username and the password shared on Slack previously) along with the most often used R packages (including the ones we used for stream processing, egbotor
,AWR.Kinesis
and thebinancer
package)de3
EC2 IAM role with full access to Kinesis, Dynamodb, Cloudwatch and theslack
token in the Parameter Storede3
security group with open ports for RStudio Server and Jenkins- lecture and seminar notes at https://github.com/daroczig/CEU-R-prod
Make sure to clean-up your EC2 nodes, security groups, keys etc created in the past weeks, as left-over AWS resources will contribute negative points to your final grade! E.g. the EC2 node you created on the second week should be terminated.
-
Minimal project (for grade up to "B"): schedule a Jenkins job that runs every hour getting the past hour's 1-minute interval klines data on ETH prices (in USD). The job should be configured to pull the R script at the start of the job either from a private or public git repo or gist. Then
- Find the min and max price of ETH in the past hour, and post these stats in the
#bots-bots-bots
MS Teams channel. Make sure to set your username for the message, and use a custom emoji as the icon. - Set up email notification for the job when it fails.
- Find the min and max price of ETH in the past hour, and post these stats in the
-
Recommended project (for grade up to "A"): Deploy an R-based API in ECS (like we did on the last week) for analyzing recent Binance (or any other real-time) data. The API should include at least 4 endpoints using different serializers, and these endpoints should be other than the ones we covered in the class. At least one endpoint should have at least a few parameters. Build a Docker image, push it to ECR, and deploy as service in ECS. Document the steps required to set up ECR/ECS with screenshots, then delete all services after confirming that everything works correctly.
Regarding feedback: by default, I add a super short feedback on Moodle as a comment to your submission (e.g. "good job" or "excellent" for grade A, or short details on why it was not A). If you want to receive more detailed feedback, please send me an email to schedule a quick call. If you want early feedback (before grading), send me an email at least a week before the submission deadline!
-
Create a PDF document that describes your solution and all the main steps involved with low level details: attach screenshots (including the URL nav bar and the date/time widget of your OS, so like full-screen and not area-picked screenshots) of your browser showing what you are doing in RStudio Server, Jenkins, in the AWS dashboards, or example messages posted in MS Teams, and make sure that the code you wrote is either visible on the screenshots, or included in the PDF.
-
STOP the EC2 Instance you worked on, but donβt terminate it, so I can start it and check how it works. Note that your instance will be terminated by me after the end of the class.
-
Include the
instance_id
on the first page of the PDF, along with your name or student id. -
Upload the PDF to Moodle.
Midnight (CET) on May 31, 2025.
An introduction to stream processing with AWS Kinesis and R: https://github.com/daroczig/CEU-R-prod/raw/2017-2018/AWR.Kinesis/AWR.Kinesis-talk.pdf (presented at the Big Data Day Los Angeles 2016, EARL 2016 London and useR! 2017 Brussels)
This section describes how to set up a Kinesis stream with a few on-demand shards on the live Binance transactions read from its websocket -- running in a Docker container, then feeding the JSON lines to Kinesis via the Amazon Kinesis Agent.
-
Start a
t3a.micro
instance running "Amazon Linux 2 AMI" (where it's easier to install the Kinesis Agent compared to using eg Ubuntu) with a known key. Make sure to set a name and enable termination protection (in the instance details)! Use SSH, Putty or eg the browser-based SSH connection. Note that the default username isec2-user
instead ofubuntu
. -
Install Docker (note that we are not on Ubuntu today, but using Red Hat's
yum
package manager):sudo yum install docker sudo service docker start sudo service docker status
-
Let's use a small Python app relying on the Binance API to fetch live transactions and store in a local file:
- sources: https://github.com/daroczig/ceu-de3-docker-binance-streamer
- docker: https://cloud.docker.com/repository/registry-1.docker.io/daroczig/ceu-de3-docker-binance-streamer
Usage:
screen -RRd streamer sudo docker run -ti --rm --log-opt max-size=50m daroczig/ceu-de3-docker-binance-streamer >> /tmp/transactions.json ## "C-a c" to create a new screen, then you can switch with C-a " ls -latr /tmp tail -f /tmp/transactions.json
-
Install the Kinesis Agent:
As per https://docs.aws.amazon.com/firehose/latest/dev/writing-with-agents.html#download-install:
sudo yum install -y aws-kinesis-agent
-
Create a new Kinesis Stream (called
crypto
) at https://eu-west-1.console.aws.amazon.com/kinesis. Don't forget to tag it (Class, Owner)! -
Configure the Kinesis Agent:
sudo yum install mc sudo mcedit /etc/aws-kinesis/agent.json
Running the above commands, edit the config file to update the Kinesis endpoint, the name of the stream on the local file path:
{ "cloudwatch.emitMetrics": true, "kinesis.endpoint": "https://kinesis.eu-west-1.amazonaws.com", "firehose.endpoint": "", "flows": [ { "filePattern": "/tmp/transactions.json*", "kinesisStream": "crypto", "partitionKeyOption": "RANDOM" } ] }
Note that extra star at the end of the
filePattern
to handle potential issues when file is copy/truncated (logrotate). -
Restart the Agent:
sudo service aws-kinesis-agent start
-
Check the status and logs:
sudo service aws-kinesis-agent status sudo journalctl -xe ls -latr /var/log/aws-kinesis-agent/aws-kinesis-agent.log tail -f /var/log/aws-kinesis-agent/aws-kinesis-agent.log
-
Make sure that the IAM role (eg
kinesis-admin
) can write to Kinesis and Cloudwatch, eg by attaching theAmazonKinesisFullAccess
policy, then restart the agentsudo service aws-kinesis-agent restart
-
Check the AWS console's monitor if all looks good there as well
-
Note for the need of permissions to
cloudwatch:PutMetricData
(see examplecloudwatch-putmetrics
policy). -
Optionally set up a cronjob to truncate that the file from time to time:
5 * * * * /usr/bin/truncate -s 0 /tmp/transactions.json
-
Set up an alert in Cloudwatch if streaming stops
As the botor
package was already installed, we can rely on the power of boto3
to interact with the Kinesis stream. The IAM role attached to the node already has the AmazonKinesisFullAccess
policy attached, so we have permissions to read from the stream.
First we need to create a shard iterator, then using that, we can read the actual records from the shard:
library(botor)
botor(region = 'eu-west-1')
shard_iterator <- kinesis_get_shard_iterator('crypto', '0')
records <- kinesis_get_records(shard_iterator$ShardIterator)
str(records)
Let's parse these records:
records$Records[[1]]
records$Records[[1]]$Data
library(jsonlite)
fromJSON(as.character(records$Records[[1]]$Data))
Exercises:
- parse the loaded 25 records into a
data.table
object with proper column types. Get some help on the data format from the Binance API docs! - count the overall number of coins exchanged
- count the overall value of transactions in USD (hint:
binance_ticker_all_prices()
andbinance_coins_prices()
) - visualize the distribution of symbol pairs
A potential solution that you should not look at before thinking ...
library(data.table)
dt <- rbindlist(lapply(records$Records, function(record) {
fromJSON(as.character(record$Data))
}))
str(dt)
setnames(dt, 'a', 'seller_id')
setnames(dt, 'b', 'buyer_id')
setnames(dt, 'E', 'event_timestamp')
## Unix timestamp / Epoch (number of seconds since Jan 1, 1970): https://www.epochconverter.com
dt[, event_timestamp := as.POSIXct(event_timestamp / 1000, origin = '1970-01-01')]
setnames(dt, 'q', 'quantity')
setnames(dt, 'p', 'price')
setnames(dt, 's', 'symbol')
setnames(dt, 't', 'trade_id')
setnames(dt, 'T', 'trade_timestamp')
dt[, trade_timestamp := as.POSIXct(trade_timestamp / 1000, origin = '1970-01-01')]
str(dt)
for (id in grep('_id', names(dt), value = TRUE)) {
dt[, (id) := as.character(get(id))]
}
str(dt)
for (v in c('quantity', 'price')) {
dt[, (v) := as.numeric(get(v))]
}
library(binancer)
binance_coins_prices()
dt[, .N, by = symbol]
dt[symbol=='ETHUSDT']
dt[, from := substr(symbol, 1, 3)]
dt <- merge(dt, binance_coins_prices(), by.x = 'from', by.y = 'symbol', all.x = TRUE, all.y = FALSE)
dt[, value := as.numeric(quantity) * usd]
dt[, sum(value)]
Let's write an R function to increment counters on the number of transactions per symbols:
-
Get sample raw data as per above (you might need to get a new shard iterator if expired):
records <- kinesis_get_records(shard_iterator$ShardIterator)$Record
-
Function to parse and process it
txprocessor <- function(record) { symbol <- fromJSON(as.character(record$Data))$s log_info(paste('Found 1 transaction on', symbol)) redisIncr(paste('USERNAME', 'tx', symbol, sep = ':')) }
-
Iterate on all records
library(logger) library(rredis) redisConnect() for (record in records) { txprocessor(record) }
-
Check counters
symbols <- redisMGet(redisKeys('^USERNAME:tx:*')) symbols symbols <- data.frame( symbol = sub('^USERNAME:tx:', '', names(symbols)), N = as.numeric(symbols)) symbols
-
Visualize
library(ggplot2) ggplot(symbols, aes(symbol, N)) + geom_bar(stat = 'identity')
-
Rerun step (1) and (3) to do the data processing, then (4) and (5) for the updated data visualization.
-
π€¦
-
Let's make use of the next shard iterator:
## reset counters redisDelete(redisKeys('USERNAME:tx:*')) ## get the first shard iterator shard_iterator <- kinesis_get_shard_iterator('crypto', '0')$ShardIterator while (TRUE) { response <- kinesis_get_records(shard_iterator) ## get the next iterator shard_iterator <- response$NextShardIterator ## extract records records <- response$Record for (record in records) { txprocessor(record) } ## summarize symbols <- redisMGet(redisKeys('USERNAME:tx:*')) symbols <- data.frame( symbol = sub('^symbol:', '', names(symbols)), N = as.numeric(symbols)) ## visualize print(ggplot(symbols, aes(symbol, N)) + geom_bar(stat = 'identity') + ggtitle(sum(symbols$N))) }
-
So far, we used the
boto3
Python module from R viabotor
to interact with AWS, but this time we will integrate Java -- by calling the AWS Java SDK to interact with our Kinesis stream, then later on to run a Java daemon to manage our stream processing application.- πͺ First, let's install Java and the
rJava
R package:
sudo apt install r-cran-rjava
- πͺ Then the R package wrapping the AWS Java SDK and the Kinesis client, then update to the most recent dev version right away:
sudo apt install r-cran-awr.kinesis sudo R -e "withr::with_libpaths(new = '/usr/local/lib/R/site-library', install.packages('AWR', repos = 'https://daroczig.gitlab.io/AWR'))" sudo R -e "withr::with_libpaths(new = '/usr/local/lib/R/site-library', devtools::install_github('daroczig/AWR.Kinesis', upgrade = FALSE))"
- πͺ Note, after installing Java, you might need to run
sudo R CMD javareconf
and/or restart R or the RStudio Server viasudo rstudio-server restart
:/
Error : .onLoad failed in loadNamespace() for 'rJava', details: call: dyn.load(file, DLLpath = DLLpath, ...) error: unable to load shared object '/usr/lib/R/site-library/rJava/libs/rJava.so': libjvm.so: cannot open shared object file: No such file or directory
- And after all, a couple lines of R code to get some data from the stream via the Java SDK (just like we did above with the Python backend):
library(rJava) library(AWR.Kinesis) records <- kinesis_get_records('crypto', 'eu-west-1') str(records) records[1] library(jsonlite) fromJSON(records[1])
- πͺ First, let's install Java and the
-
Create a new folder for the Kinesis consumer files:
streamer
-
Create an
app.properties
file within that subfolder
executableName = ./app.R
regionName = eu-west-1
streamName = crypto
applicationName = my_demo_app_sadsadsa
AWSCredentialsProvider = DefaultAWSCredentialsProviderChain
- Create the
app.R
file:
#!/usr/bin/Rscript
library(logger)
log_appender(appender_file('app.log'))
library(AWR.Kinesis)
library(methods)
library(jsonlite)
kinesis_consumer(
initialize = function() {
log_info('Hello')
library(rredis)
redisConnect(nodelay = FALSE)
log_info('Connected to Redis')
},
processRecords = function(records) {
log_info(paste('Received', nrow(records), 'records from Kinesis'))
for (record in records$data) {
symbol <- fromJSON(record)$s
log_info(paste('Found 1 transaction on', symbol))
redisIncr(paste('symbol', symbol, sep = ':'))
}
},
updater = list(
list(1/6, function() {
log_info('Checking overall counters')
symbols <- redisMGet(redisKeys('symbol:*'))
log_info(paste(sum(as.numeric(symbols)), 'records processed so far'))
})),
shutdown = function()
log_info('Bye'),
checkpointing = 1,
logfile = 'app.log')
-
πͺ Allow writing checkpointing data to DynamoDB and CloudWatch in IAM
-
Convert the above R script into an executable using the Terminal:
cd streamer
chmod +x app.R
- Run the app in the Terminal:
/usr/bin/java -cp /usr/local/lib/R/site-library/AWR/java/*:/usr/local/lib/R/site-library/AWR.Kinesis/java/*:./ \
com.amazonaws.services.kinesis.multilang.MultiLangDaemon \
./app.properties
- Check on
app.log
-
Reset counters
library(rredis) redisConnect() keys <- redisKeys('symbol*') redisDelete(keys)
-
πͺ Install the
treemap
packagesudo apt install r-cran-httpuv r-cran-shiny r-cran-xtable r-cran-htmltools r-cran-igraph r-cran-lubridate r-cran-tidyr r-cran-quantmod r-cran-broom r-cran-zoo r-cran-htmlwidgets r-cran-tidyselect r-cran-rlist r-cran-rlang r-cran-xml r-cran-treemap r-cran-highcharter
-
Run the below Shiny app
## packages for plotting library(treemap) library(highcharter) ## connect to Redis library(rredis) redisConnect() library(shiny) library(data.table) ui <- shinyUI(highchartOutput('treemap', height = '800px')) server <- shinyServer(function(input, output, session) { symbols <- reactive({ ## auto-update every 2 seconds reactiveTimer(2000)() ## get frequencies symbols <- redisMGet(redisKeys('symbol:*')) symbols <- data.table( symbol = sub('^symbol:', '', names(symbols)), N = as.numeric(symbols)) ## color top 3 symbols[, color := 1] symbols[symbol %in% symbols[order(-N)][1:3, symbol], color := 2] ## return symbols }) output$treemap <- renderHighchart({ tm <- treemap(symbols(), index = c('symbol'), vSize = 'N', vColor = 'color', type = 'value', draw = FALSE) N <- sum(symbols()$N) hc_title(hctreemap(tm, animation = FALSE), text = sprintf('Transactions (N=%s)', N)) }) }) shinyApp(ui = ui, server = server, options = list(port = 3838))
We will learn more about Shiny in the upcoming Data Visualization 4 class :)
Will be updated from week to week.
File a GitHub ticket.