IN PERSON: Apache Kafka® Meetup - Sept 2024

Join us for an Apache Kafka® meetup on Tuesday, September 24th from 5:30pm hosted by Dynatrace in Vienna!

📍Venue:
Dynatrace
The Icon Vienna, Tower 17, 17th Floor, Wiedner Gürtel 13, 1100 Wien, Austria

Public transport
We're located in The Icon Vienna center, next to Vienna Central Train Station.
Option 1: Take U1 subway line.
Option 2: Take S1, S2, S3, S4, S60, or S80.
Option 3: Take a tram D, 0 or 18.
Option 4: Take a bus 13A or 69A.

Parking information
The Icon Vienna garage is located directly under the building and offers parking for visitors. The prices are available on the website of The Icon Vienna, but you need to pay for it without a refund from Dynatrace

🗓 Agenda:

• 5:30pm: Doors open
  5:30pm – 6:00pm: Drinks and networking
  6:00pm - 6:30pm: Mathias Frey, Head of Platform Architecture at Erste Group Bank AG
  6:30pm - 7:00pm: Igor Pernek, Team Lead & Principal Software Engineer, Dynatrace
• 7:00pm - 8:00pm: Additional Q&A and Networking, Snacks/Drinks

💡 Speaker One:
Mathias Frey, Head of Platform Architecture at Erste Group Bank AG

Title of Talk:
Speeding up analytics(mis)using Kafka

Abstract:
Real-time is one of the hot buzzwords in our industry. Yet, when it comes to large scale, complex tasks, an array of hard-to-manage source systems, and zero tolerance for error, we’re pushed back to processing data in more conventional ways. In my talk, I will present how we (try to) overcome the issue at Erste Group -- showing the good, the bad and the ugly of real-time analytics.

Bio:
More than twenty years of experience in software development and related functions, both in small companies as well as large enterprises. As Head of Platform Architecture at Erste Group responsible for tech strategy of everything that eventually faces our customers. Intrigued by reactive programming.

💡 Speaker Two:
Igor Pernek, Team Lead & Principal Software Engineer, Dynatrace

Title of Talk:
Horizontally scalable query scheduling with Kafka

Abstract:
A common challenge for applications doing distributed processing is finding the best node to execute a particular workload. In generic use cases, off-the-shelf load balancers are hooked with resource utilisation metrics and used to drive node selection. However, more complicated scenarios require implementing custom business logic to provide additional scheduling features such as: retries on backpressure or processing failures, node selection based on workload cost estimation, rescheduling of workloads in progress to more suitable nodes, etc.

To be able to support those scenarios for query scheduling, we build on top of the event-driven architecture pattern and utilize Kafka to implement a distributed query scheduler that is highly available, horizontally scalable, and guarantees exactly-once semantics for query execution. In this talk, we will sketch the high-level architecture of the approach and provide interesting low-level details.

Bio:
Igor is a Principal developer and team lead at Dynatrace. He is currently working on Grail, a highly-available distributed database that powers 3rd generation of Dynatrace services. Before Dynatrace he was tackling large scale data processing problems, such as processing thousands of requests per second with sub-second latency. He also holds a PhD in Computer Science, specialising in applying machine learning algorithms on wearable sensor data.

***
DISCLAIMER
BY ATTENDING THIS EVENT IN PERSON, you acknowledge that risk includes possible exposure to and illness from infectious diseases including COVID-19, and accept responsibility for this, if it occurs.
NOTE: We are unable to cater for any attendees under the age of 21.
***
If you would like to speak or host our next event please let us know! community@confluent.io