Bengaluru Systems Meetup #5

Apache Kafka is the industry’s de facto standard implementation of a distributed log. Hidden behind the innocuous Kafka API are the nitty-gritty details of organising data on the machine to satisfy the requirements of durability, fault tolerance, and cost-effectiveness while supporting high read and write throughputs.

We will try to do a whirlwind tour of open-source Kafka’s storage model and the decisions that have gone into building it out. We then extend beyond the basics to learn how other modern Kafka-compatible systems like Redpanda, Warpstream, Kora etc. tackle these problems.

I hope folks leave the talk with a new-found appreciation of system internals and stay ever more curious and excited about all the goodness ahead...