Is Your Kafka Data in Working Order? Tactics for Overcoming Kafka Data Order Issues


An incomplete understanding of how Apache Kafka stores information can mean hanging an “Out of Order” sign over your application and scrambling to appease upset customers – and all because retrieved data is literally out of the sequence your application needs it to be in. While Kafka is a highly-capable open source platform for capturing streaming data, wrangling that information to avoid data order issues and receive that data precisely as you require it can take some specific expertise.

The SCALE 18X audience for this session can expect to come away with an in-depth understanding of how Kafka stores and retrieves streaming data, including how to organize data and preserve system balance by utilizing Kafka topic partitions and keying. Attendees will also learn to implement best practices and leverage techniques including Exactly-Once Message Delivery to instill system-wide resilience, overcome potential failures, and prevent issues before they can occur. For attendees for whom data is the lifeblood of their businesses (i.e. almost all businesses!), the insights offered during this presentation will enable more successful and reliable enterprise-grade Kafka deployments.

The audience will leave understanding how to resolve Kafka issues using a combination of

-- Effective topic and partititioning strategies

-- Effective data keying strategies

-- Exactly-once semantics with Producers and Consumers

Room 104
Sunday, March 8, 2020 - 15:00 to 16:00