Updated on 2024-06-07 GMT+08:00

Table Partitioning

Table partitioning logically divides a large table or index into smaller and easier-to-manage logical units (partitions), minimizing the impact on table query and modification statements. Users can quickly locate a partition where data is located by using a partition key. In this way, users do not need to scan all large tables in the database and can concurrently perform DDL and DML operations on different partitions. Table partitioning provides users with the following capabilities:
  1. Improve query efficiency in large-capacity data scenarios: Because data in a table is logically partitioned by partition key, the query result can be implemented by accessing a partition subset instead of the entire table. This partition pruning technique can provide an order of magnitude performance gain.
  2. Reduce the impact of parallel O&M and query operations. The mutual impact of parallel DML and DDL statements is reduced, which is obvious in scenarios where a large amount of data is partitioned by time. For example, new data partitions are imported to the database and queried in real time, and old data partitions are cleaned and merged.
  3. Provide flexible data O&M management in large-capacity scenarios: Partitioned tables physically isolate data in different partitions at the table file level. Each partition can have independent physical attributes, such as data compression, physical storage settings, and tablespaces. In addition, it supports data management operations, such as data loading, index creation and rebuilding, and partition-level backup and restoration, instead of performing operations on the entire table, reducing operation time.