As a part of the overall feature and functionality updates to the Ascend Platform, we’ve released a number of improvements and features.

:rocket: FEATURES :rocket:

  • Spark SQL GA. This enables the full power of spark sql, including all spark supported types, functions, subqueries, all backed by Ascend’s partitioning mechanism (available under advanced settings)
  • Add Read Connector UI support for byte, binary, short and decimal types
  • Support Spark 3.1.2 runtime
  • Add CDC for MySQL Read Connector

:sparkles: ENHANCEMENTS :sparkles:

  • MsSQL Read Connector: reset cursor when a LSN (log sequence number) reset occurs

:wrench: BUGFIXES :wrench:

  • Fix OutOfMemory errors that occasionally cause scheduler downtime
  • JDBC Read Connectors: quote column names when running queries for incremental pulls
  • Apply a patch to the default Spark runtime to avoid a pod cycling edge case in the JDBC server / query server. Spark issue: [SPARK-33711][K8S] Avoid race condition between POD lifecycle manager and scheduler backend

If you need support, have questions, or would like to get started on the platform, schedule time here!