Data Partition O&M Management

A partitioned table provides flexible support for data lifecycle management (DLM). DLM is a set of processes and policies used to manage data throughout the service life of data. An important component is to determine the most appropriate and cost-effective medium for storing data at any point in the data lifecycle. New data used in daily operations is stored on the fastest and most available storage tier, while old data that is infrequently accessed may be stored on a less costly and inefficient storage tier. Old data may also be updated less frequently, so it makes sense to compress the data and store it as read-only.

Partitioned tables provide an ideal environment for implementing the DLM solution. Different partitions use different tablespaces, maximizing usability and reducing costs in the data lifecycle. The settings are performed by database O&M personnel on the server. Actually, users are unaware of the optimization settings. Logically, users still query the same table. In addition, O&M operations, such as backup, restoration, and index rebuilding, can be performed on different partitions. The Divide and Conquer method is implemented on different subsets of a single dataset to meet differentiated requirements of service scenarios.