This topic describes the definition and scenarios of a routing algorithm.


A routing algorithm is used to identify physical tables or table shards that are mapped to a logical table in a logical database based on routing fields. You can configure a routing algorithm for a logical database to precisely identify intended physical tables based on specified conditions. This reduces the overhead during routing and improves the efficiency of your data operations.
Note If no routing algorithm is configured, operations on the data of a logical table traverse all the physical tables that correspond to the logical table. This significantly prolongs the time required for data processing.

A routing algorithm consists of routing fields and an algorithm.


  • Query the data of table shards.
  • Change the data of table shards.
  • Export the data of table shards.


You can find the desired logical table in the logical database, and modify its routing algorithm.