Cloudera Streaming Analytics 1.6 Launch Notes

We’re excited to announce the discharge of Cloudera Streaming Analytics (CSA) 1.6 for CDP Non-public Cloud Base. With this launch, we construct on the inspiration on 1.4 and 1.5 – with quite a lot of fixes, enhancements, and options. Beginning with this launch, we now have an aligned launch cycle for CSA Neighborhood Version (CE). Now you can anticipate simultaneous releases of CSA for each CE and CDP Non-public Cloud Base variations. This may make sure you get your palms on the most recent options first, and we hope you’ll be able to give us suggestions early and infrequently.

Cloudera SQL Stream Builder was initially launched in CSA 1.3. Since then, we now have seen nice traction and quite a lot of manufacturing implementations spanning from medium to extraordinarily massive in dimension. We’ve been capturing buyer suggestions, and have integrated it into this launch. A few of these enhancements and options are:

  • Flink JAR submission (for Java UDF’s)
  • Logging enhancements throughout the board
  • DB2 Change Knowledge Seize (CDC) and JDBC connectivity
  • RHEL 8.x compatibility
  • Flink 1.14
  • JDBC set up directions for CE
  • Safety enhancements (addresses CVE-2021-44228)
  • Inner optimizations and enhancements that allow quicker CSA growth

You may see detailed launch notes within the documentation.

CSA CE has been launched since model 1.5 and the suggestions has been unimaginable. However we need to deal with one query that retains developing – does CSA CE make sense as my most important growth surroundings for stream processing jobs? The reply is, primarily, sure! Historically, Cloudera has launched trial variations of CSA software program. However, CE fully removes the necessity for the trial model – you possibly can check out CSA to your coronary heart’s content material or till your POC is full. Nonetheless, CE goes even additional, and is sensible to make use of as your everlasting growth surroundings! 

The workflow we anticipate is one thing like this:

  • Compose SQL, and construct jobs/processors utilizing CSA CE
  • Run in your desktop or cloud node, connecting to Kafka or different sources/sinks through API calls to their respective clusters.
  • Run/check/iterate on the CE surroundings, till your job is prepared for manufacturing.
  • Save your SQL, UDF’s, and so forth into recordsdata (maybe in a supply code repository) and run/handle it through REST on manufacturing variations of CSA (once more through API calls).

We hope this helps clear up some questions round what CSA CE can be utilized for and recommended configurations and architectures. We plan future weblog posts on this workflow. Till then, if in case you have questions or suggestions you possibly can all the time hit up the staff at