A small bit of Druid magic in large Data planet • Michał Misiewicz • Devoxx Poland 2021

IT planet is changing rapidly, especially around large Data. We have started simply, by providing services to our customers. Later we understood the request of analytic platforms, based on which, business decisions are made. Hadoop stack became a first viable solution and it was good adequate for the time being. However, throughout fresh years requirements have greatly increased. Now we request data driven systems, which are reacting to changing planet in a real time manner. The emergence of IoT systems became a real challenge even to state of the art large Data stacks. Our production systems make dozens of thousands of events per seconds. The crucial question is: how to analyse this immense data stream in an interactive manner? - with a small bit of Druid magic. This talk is simply a method dive into Apache Druid - data warehouse which revolutionized the way users research immense datasets.
First part of my presentation will cover method aspects of Apache Druid. Later I want to share my 3 years of experience in creating analytic platforms. At the end I am going to show Apache Druid in action by presenting a live demo.

The talk will cover:
-Highly distributed and scalable architecture
-Data modeling
-Druid as a heart of modern data analytics platform
-Casting spells: make large Data tiny again!
-Live demo

Lecture took place on Friday 27th August 2021 at 11:40 in area 2

Michał Misiewicz - chief technology officer at Datumo. Designer of analytic platforms based on Apache Druid. Open origin projects contributor - Apache Druid, Apache Airflow and Apache NiFi. In free time he is simply a runner.

#IT #Development #SoftwareDevelopment