Scenario Overview

One of the purposes of row compression is to reuse the space saved after compressing tables. The process can be summarized as follows:

• Compression execution. The system traverses each page of tables with Information Lifecycle Management (ILM) enabled and performs compression. The compressed data is referred to as BCA and remains stored in the original page. After compression, the page has more available space. Note that the space saved after compression is not released back to the operating system. For details about ILM, see Data Lifecycle Management: OLTP Table Compression.
• Data insertion. When new data is inserted, the system prioritizes pages with enough remaining space to accommodate the new rows. By increasing the amount of free space available in pages, row compression reduces the actual storage occupied by the table.

The following examples demonstrate how row compression can be scheduled and executed:

• Manual scheduling: You need to manually call the compression interface, and only one task can be generated at a time.
• Automatic scheduling: After the prerequisite configuration is complete, the system automatically creates compression tasks in the background through scheduled jobs, and one scheduling cycle can generate multiple tasks.