Application Scenarios

The OmniOperator feature is mainly used in data analytics engines. It optimizes the execution process to improve data analytics performance.

OmniOperator provides native operators that replace the open source operators in analytics engines, accelerating SQL queries and enhancing data analytics performance. After a user submits an SQL statement, the engine converts the SQL statement into a series of operators. OmniOperator replaces some of the open source operators with native operators to improve the execution efficiency.

OmniOperator is suited for large-scale data fusion analytics scenarios and can effectively handle high-concurrency and large-volume data processing requirements.

OmniOperator supports the following analytics engines:

• Spark adaptation frameworks
  • SparkExtension: Spark 3.1.1, 3.3.1, 3.4.3, and 3.5.2
  • Gluten: Spark 3.3.1, which must run on a Kunpeng server that supports the Scalable Vector Extension (SVE) instruction set.
• Hive adaptation framework
  • HiveExtension: Hive 3.1.0