We are excited to announce the release of Cloudera Streaming Analytics (CSA) 1.6 for CDP Private Cloud Base. With this release, we build on the foundation on 1.4 and 1.5 – with a number of fixes, enhancements, and features. Starting with this release, we now have an aligned release cycle for CSA Community Edition (CE). You can now expect simultaneous releases of CSA for both CE and CDP Private Cloud Base versions. This will ensure you get your hands on the newest features first, and we hope you are able to give us feedback early and often.
Cloudera SQL Stream Builder was initially released in CSA 1.3. Since then, we have seen great traction and a number of production implementations spanning from medium to extremely large in size. We’ve been capturing customer feedback, and have incorporated it into this release. Some of these improvements and features are:
- Flink JAR submission (for Java UDF’s)
- Logging improvements across the board
- DB2 Change Data Capture (CDC) and JDBC connectivity
- RHEL 8.x compatibility
- Flink 1.14
- JDBC install instructions for CE
- Security improvements (addresses CVE-2021-44228)
- Internal optimizations and improvements that enable faster CSA development
You can see detailed release notes in the documentation.
CSA CE has been released since version 1.5 and the feedback has been incredible. But we want to address one question that keeps coming up – does CSA CE make sense as my main development environment for stream processing jobs? The answer is, essentially, yes! Traditionally, Cloudera has released trial versions of CSA software. But, CE completely removes the need for the trial version – you can try out CSA to your heart’s content or until your POC is complete. However, CE goes even further, and makes sense to use as your permanent development environment!
The workflow we anticipate is something like this:
- Compose SQL, and build jobs/processors using CSA CE
- Run on your desktop or cloud node, connecting to Kafka or other sources/sinks via API calls to their respective clusters.
- Run/test/iterate on the CE environment, until your job is ready for production.
- Save your SQL, UDF’s, etc into files (perhaps in a source code repository) and run/manage it via REST on production versions of CSA (again via API calls).
We hope this helps clear up some questions around what CSA CE can be used for and suggested configurations and architectures. We plan future blog posts on this workflow. Until then, if you have questions or feedback you can always hit up the team at https://community.cloudera.com.