December 2016 Summaries

The text discusses microservices architecture, emphasizing on the importance of data sharing between services in a distributed system. It highlights the challenges faced when sharing significant datasets between services and suggests using a Distributed Log for this purpose. The author proposes using a Streaming Platform like Kafka to enable stateful stream processing within each consuming service, allowing them to join and operate on shared streams while keeping data true to its source. This approach helps in building more scalable, fungible, and autonomous services that can survive and evolve for decades.

Confluent Platform now offers Multi-Datacenter Replication, a feature that automates data replication between Kafka clusters across multiple datacenters, making it easy to manage data flow and placement without writing code. This feature is part of Confluent Platform 7.7 and provides hassle-free replication at scale, with features such as automatic topic creation and modification, fault tolerance, and monitoring through Confluent Control Center. The feature can be used for various use cases, including aggregation and analytics, and can be easily implemented using the Kafka Connect API.

Apache Kafka 3.8.0 has been released, bringing numerous new features and improvements to the platform, which are detailed in the release notes. This update coincides with various developments in the streaming community and the introduction of KIPs that promise to significantly influence Kafka's semantics and operability. Confluent has also launched version 3.1 of its Confluent Platform. The community humorously addresses the perennial "tabs versus spaces" debate by utilizing the new confluent-kafka-javascript client to create an application that tracks and displays vote counts on this topic using Kafka and a JavaScript frontend.