Rate This Document
Findability
Accuracy
Completeness
Readability

Change Description

The OmniRuntime OmniOperator feature of Kunpeng BoostKit for Big Data uses one base to support different engines (such as Spark), reducing repeated optimization work, fully exploring common and heterogeneous computing power, and promoting the Kunpeng ecosystem.

New Features

  • Support for Spark 3.4.3 and Spark 3.5.2.
  • Spark 3.4.3 and Spark 3.5.2 support the LIMIT...OFFSET... syntax.
  • New expressions: try_add, try_divide, try_multiply, try_subtract, try_avg, and try_sum
  • Spark 3.5.2 supports the join type and build side co-directional logic of the native ShuffledHashJoin operator.
  • Added the WindowGroupLimit operator for Spark 3.5.2 to optimize the ROW_NUMBER and RANK functions for Top-K calculation.

Modified Features

None

Removed Features

This version does not adapt to CentOS 7.9.