Creating an independent column-organized, partitioned environment

Introduce a columnar workload to co-exist with, but not interact with, existing row applications by loading a new dataset into the database.

Before you begin

• Review the best practices for column-organized table configuration
• To minimize the impact of the configuration changes on the row-table applications, set the DEGREE value for the row-table applications to 1 and the column-table applications to ANY. Use the MAXIMUM_DEGREE attribute to achieve this.

About this task

Create new column-organized tables in the database, then load the desired information into them.

To introduce independent columnar workloads to the database:

1. Create a table by running the CREATE TABLE command with the ORGANIZE BY COLUMN option:

   CREATE TABLE <new_table_name> (
    <column_1> <datatype>,
    <column_2> <datatype>,
    <column_3> <datatype>)
    ORGANIZE BY COLUMN;

2. Load data into the table using the INGEST, INSERT, IMPORT, or LOAD command:

   LOAD FROM <file_with_data> of <file_with_data's_directory> REPLACE INTO <new_table_name>