Beyond the Basics: Building Resilient Async Systems with Kafka
Details
As a community, we always share knowledge with a prime motive of helping the community to overcome daily challenges that Technologists face; in this process, we learn as well.
Summary:
Apache Kafka is the backbone of modern distributed systems, but using it effectively requires more than just knowing how to publish a message. In this intermediate-to-advanced session, we move beyond “Hello World” to explore the architectural patterns and operational realities of running Kafka in production.
We will dissect the core principles of asynchronous design—from enforcing strict ordering via semantic partitioning to achieving “exactly-once” processing semantics. We will discuss how to navigate the critical trade-offs between the plain Consumer API and Kafka Streams, and how to manage state. In addition, there will be a discussion on “Day 2” operational challenges, demonstrating how to eliminate “stop-the-world” rebalancing storms and how to future-proof your cluster with tiered storage. The speakers will also draw from their experiences on how they used these principles to recover data when they encountered a data corruption in production.
This will be for:
Backend Engineers, Architects, and FullStack professionals who want to harden their async architecture.
Presenters:
Mahesh Lal has 16 + years of experience, has built and led high-performing engineering teams delivering complex, impact-driven applications at massive scale.
Currently a consultant serving as Principal at Technohise, he focuses on improving and modernising critical systems.
He is also the author of Neo4j Graph Data Modeling, a book that introduces the core concepts of graph databases and effective data modelling.
Zoom:
https://us06web.zoom.us/j/82188499974?pwd=8KDnaZn5Rh60ki7co7autbaGKULrtp.1
SPOC: Vandana
Call: +91 96201 39111