V1.4.0

Change Description

The OmniShuffle shuffle acceleration feature of Kunpeng BoostKit for Big Data uses the plugin mechanism provided by Spark to implement the Shuffle Manager and Broadcast Manager plugin interfaces and replace the native Shuffle and Broadcast of Spark in a non-intrusive manner. It reduces disk I/Os, accelerates data exchange between nodes, and improves query efficiency.

Version 1.4.0 has the following changes:

Optimized NEON instructions, mainly for key read/write processes in the memory.

Resolved Issues

None

Known Issues

None