SET/RESET
Function
This command is used to dynamically add, update, display, or reset Hudi parameters without restarting the driver.
Syntax
- Add or update a parameter value:
SET parameter_name=parameter_value
This command is used to add or update the value of parameter_name.
- Display a parameter value:
This command is used to display the value of parameter_name.
- Display session parameters:
This command is used to display all supported session parameters.
- Display session parameters along with usage details:
This command is used to display all supported session parameters and their usage details.
- Reset parameter values:
This command is used to reset all session parameters.
Parameter Description
Parameter |
Description |
---|---|
parameter_name |
Name of the parameter to be dynamically added, updated, or displayed. |
parameter_value |
New value to be set for parameter_name. |
Precautions
The following table lists the properties to be used in the SET or RESET commands.
Property |
Description |
---|---|
hoodie.insert.shuffle.parallelism |
Degree of parallelism (DOP) of Spark shuffle for writing data in insert mode. |
hoodie.upsert.shuffle.parallelism |
DOP of Spark shuffle for writing data in upsert mode. |
hoodie.delete.shuffle.parallelism |
DOP of Spark shuffle for deleting data in delete mode. |
hoodie.sql.insert.mode |
Insert mode. The value can be strict, non-strict, or upsert. |
hoodie.sql.bulk.insert.enable |
Whether to enable bulk insert. |
spark.sql.hive.convertMetastoreParquet |
Converts the parquet table into a data source table for reading. If the provider of Hudi is Hive and Spark SQL or Spark Beeline is used to read data, set this parameter to false. |
Examples
- Add or Update command:
set hoodie.insert.shuffle.parallelism = 100; set hoodie.upsert.shuffle.parallelism = 100; set hoodie.delete.shuffle.parallelism = 100;
- Reset command:
RESET
System Response
- You can view the success result in driver logs.
- You can view the failure result on the UI.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.