Boosting Kafka Performance in a Day

We’ve all tried adjusting the number partitions, replica fetch size, batch size, socket buffer size, number of I/O threads … you could spend an hour enumerating the tuning options for Kafka to squeeze a few percent of extra performance. Finding the right combination of parameters is closer to alchemy than engineering work, time consuming and in bigger deployments sometimes even impossible!
What if there's a better way? What if you could "just" make Kafka more efficient and gain better performance without months of engineering effort?
Since Kafka is written in Java and Scala, in this talk, we'll discuss the effect of Java Virtual Machine (JVM) on a Kafka cluster. We'll demonstrate that running Kafka on Azul Prime JVM can reduce the latency by up to 40% while handling the same load. We’ll go beyond shiny marketing-y charts, dig into flamegraphs, and look at assembly code to fully understand where the speed up is coming from.
After the session, you'll understand the importance of the underlaying JVM and how you can leverage this knowledge to boost the performance of the cluster to achieve better SLAs or reduce the infrastructure costs.
ABOUT CONFLUENT
Confluent is pioneering a fundamentally new category of data infrastructure focused on data in motion. Confluent’s cloud-native offering is the foundational platform for data in motion - designed to be the intelligent connective tissue enabling real-time data, from multiple sources, to constantly stream across the organization. With Confluent, organizations can meet the new business imperative of delivering rich, digital front-end customer experiences and transitioning to sophisticated, real-time, software-driven backend operations. To learn more, please visit www.confluent.io.
#confluent #apachekafka #kafka