What's new in Cloudera Streaming Analytics

Cloudera Streaming Analytics 1.17.0 covers new features beside the core streaming functionality of Apache Flink and SQL Stream Builder.

Cloudera platform support

Cloudera Streaming Analytics 1.17.0 is supported on Cloudera on premises 7.3.2.0. Review the 7.3.2.0 Release Notes and Support Matrix to understand which operating system, database, and JDK versions are supported for Cloudera Streaming Analytics as well.

Framework and UI upgrades

The underlying frameworks for Cloudera SQL Stream Builder are upgraded to improve platform stability and resolve vulnerabilities.

Enhanced Kafka Data Source configurations

You can specify custom properties for Kafka connections directly in the Kafka Data Source form. Additionally, you can explicitly define the truststore and keystore types (JKS, PKCS12, or BCFKS) when configuring a secured Kafka cluster.

Removal of JavaScript UDFs

All JavaScript User-Defined Function (UDF) features have been removed from Cloudera Streaming Analytics. Cloudera recommends rewriting your existing UDFs in Python before upgrading to Cloudera Streaming Analytics 1.17.0.

For more information about working with Python UDFs, see Adding Python UDFs.

Removal of Data Transformation for Kafka tables

Data Transformation for Kafka tables (also referred to as input transforms) relied on JavaScript and is supported only on Java Development Kit (JDK) 11 and lower versions. Cloudera Streaming Analytics 1.17.0 runs on Java 17, so Data Transformation for Kafka tables has been removed, similar to JavaScript User-Defined Functions (UDFs).

Cloudera recommends expressing equivalent logic in Flink SQL or by using Python UDFs.

For more information about working with Python UDFs, see Adding Python UDFs.