Rate This Document
Findability
Accuracy
Completeness
Readability

Change Description

The OmniStream Flink Native feature of Kunpeng BoostKit for Big Data uses native code (C/C++) to implement open source operators, accelerating the Flink engine execution and improving query performance.

New Features

SQL

  • Accelerated the Calc, GroupAgg, Join, Deduplicate, Rank, Window, and Kafka Source/Sink operators.
  • Provided an efficient data organization mode, OmniVec.
  • Added support for memory and RocksDB status backends.

DataStream

  • Accelerated the Kafka Source, Kafka Sink, Map, FlatMap, Reduce, and filter operators.
  • Implemented the basic UDF framework and basic UDF translation library, and enabled the UDF automatic native framework to run DataStream Wordcount stateful and stateless cases.
  • Added support for the memory status backend.

Modified Features

None

Removed Features

None