Chapter12

Gathering Insights Through Data Mining

This exercise shows a simple Apache Flink job performing feature extraction from traces.

Full instructions are in the book, but as a quick summary:

• docker-compose up brings up the Jaeger backend, Kafka, Elasticsearch, and Kibana
• the main Java program runs the Flink job that stores extracted features back into Elasticsearch
• the sample traces are generated with microsim simulator
• resultes can be viewed/plotted in Kibana