Event Streaming And Real Time Data Processing

Discover how Globex utilizes event streaming to deliver personalized product recommendations and insights. 

Enhance E-commerce with Real-Time Personalization

2
●
Globex
-
A
fictitious
retail
e-commerce
company
wants
to
enhance
their
online
store
application
and
provide
a
real
time
personalized
experience
for
its
customers.
●
●
They
wanted
track
user
activity
and
gain
business
insights
on
which
products
are
most
popular
and
provide
personalized
recommendations
to
the
users
2
Building
Featured
Products
and
Recommendations
with
Event
Streaming
Red
Hat
OpenShift
GitOps

Untitled presentation - Google Slides

Event and data streaming platforms, such as Apache Kafka, enable applications to process and react to events individually, enabling real-time data processing, storage, analysis, and scalability.

Kafka
Topics
Event
Streaming
And
Real
Time
Data
Processing
AMQ
streams,
Quarkus
Track
user
activity
Analyse
real
time
with
Kafka
Streams
and
display
Featured
Products

[External] Cloud Native Architecture Workshop Overview - Google Slides

A
globex-app-user1
A
kafdrop
A
strimzi-kafka
K
kafka
D
activi…acking
D
recomm…engine
D
catalog
D
inventory
D
activi…ulator
D
invent…tabase
D
catalog-database
D
kafdrop
D
kafka-…erator
D
globex-ui

Topology · Red Hat OpenShift

A
kafdrop
A
strimzi-kafka
K
kafka
D
kafdrop
A
globex-app-user1
D
kafka-…erator
D
activi…acking
D
recomm…engine
D
catalog
D
inventory
D
activi…ulator
D
invent…tabase
D
globex-ui
D
catalog-database

A walkthrough of the application that leverages this architecture to display a list of featured products, based on user activity that is streamed to a kafka topic

Let's explore the Globex Website

A
globex-app-user1
D
catalog
D
invent…tabase
D
catalog-database
D
inventory
D
activi…acking
A
strimzi-kafka
K
kafka
D
recomm…engine
D
activi…ulator
A
kafdrop
D
kafka-…erator
D
kafdrop
D
globex-ui

Globex Online

This Kafka topic will receive the user activity events generated on the Globex web site.

Kafka Topic Creation

Import YAML · Red Hat OpenShift

A
globex-app-user1
A
kafdrop
A
strimzi-kafka
K
kafka
D
activi…acking
D
recomm…engine
D
globex-ui
D
catalog
D
inventory
D
activi…ulator
D
invent…tabase
D
catalog-database
D
kafka-…erator
D
kafdrop

Kafdrop: Broker List

To connect to a Kafka broker, the application needs to be configured with connection details like the Kafka bootstrap server URL, the security protocol and the user credentials. These details are stored in a secret

Binding applications to Apache Kafka

Secrets · Red Hat OpenShift

kafka-client-secret · Secret · Details · Red Hat OpenShift

ServiceBinding
SB
kafka-servicebinding
Connected
Actions

kafka-servicebinding · Service Binding · Details · Red Hat OpenShift

Volumes
Name
	
Mount path
	
SubPath
	Type	
Permissions
	
Utilized by
	
config	/deployments/config	No subpath	
Secret
S
activity-tracking
	Read/Write	activity-tracking	

kafka-servicebinding	/bindings/kafka-servicebinding	No subpath	
Secret
S
kafka-client-secret
	Read/Write	activity-tracking

activity-tracking · Deployment · Details · Red Hat OpenShift

D
kafdrop
A
kafdrop
A
strimzi-kafka
K
kafka
A
globex-app-user1
D
kafka-…erator
D
globex-ui
D
catalog
D
invent…tabase
D
inventory
D
catalog-database
D
activi…ulator
D
recomm…engine
D
activi…acking

A
kafdrop
A
strimzi-kafka
K
kafka
D
kafdrop
A
globex-app-user1
D
kafka-…erator
D
globex-ui
D
catalog
D
inventory
D
activi…ulator
D
invent…tabase
D
catalog-database
D
activi…acking
D
recomm…engine

Deployment
D
recommendation-engine
Actions
Details
Resources
Observe
1Pod
Name
recommendation-engine
Namespace
Namespace
NS
globex-user1
Labels
Edit
app.kubernetes.io/instance
=
globex-app-user1
app.kubernetes.io/managed-by
=
Helm
app.kubernetes.io/name
=
recommendation-engine
app.kubernetes.io/part-of
=
globex-app-user1
app.kubernetes.io/version
=
1.0.0
app.openshift.io/runtime
=
quarkus
helm.sh/chart
=
recommendation-engine-0.1.0
service-binding/type
=
kafka
Pod selector
app.kubernetes.io/name=recommendation-engine, app.kubernetes.io/part-of=globex-app-user1
Node selector

No selector

Tolerations
0 tolerations
Annotations
2 annotations
Status
Active
Created at
Sep 3, 2024, 3:41 PM
Owner
No owner
Update strategy
RollingUpdate
Max unavailable
25% of 1 pod
Max surge
25% greater than 1 pod
Progress deadline seconds
600 seconds
Min ready seconds
Not configured
PodDisruptionBudget
No PodDisruptionBudget
VerticalPodAutoscalers
No VerticalPodAutoscalers

A
globex-app-user1
D
activi…acking
D
globex-ui
D
catalog
D
invent…tabase
D
inventory
A
kafdrop
D
activi…ulator
D
catalog-database
A
strimzi-kafka
K
kafka
D
kafka-…erator
D
recomm…engine
D
kafdrop

Now that the Activity Tracking and Recommendation Engine apps are up and running, we can test the generation of activity events and the calculation of the top featured products

Testing the Globex Coolstuff application

A
kafdrop
A
strimzi-kafka
K
kafka
D
kafdrop
A
globex-app-user1
D
kafka-…erator
D
activi…acking
D
globex-ui
D
catalog
D
inventory
D
invent…tabase
D
catalog-database
D
recomm…engine
D
activi…ulator

OpenAPI UI (Powered by Quarkus 2.13.3.Final)

Kafdrop: Topic: globex.tracking

Kafdrop: Topic: globex.tracking: Messages

globex.recommendation-product-score-aggregated-changelog

Kafdrop: Topic: globex.recommendation-product-score-aggregated-changelog

Kafdrop: Topic: globex.recommendation-product-score-aggregated-changelog: Messages

Learn more about Red Hat's Kafka Offering

 We've now successfully built an event streaming platform using Red Hat Streams for Apache kafka to track user activity on an ecommerce website to find out the most popular products in the store and build a recommendation engine. 

We hope you found this walkthrough useful. Below are some resources that can help you get started with your event streaming and real time data processing journey