📣
TiDB Cloud Premium is now in public preview. Unlimited growth, instant elasticity, advanced security for enterprise workloads. Try it out →

CLUSTERING_INFORMATION



Returns clustering information of a table.

Syntax

CLUSTERING_INFORMATION('<database_name>', '<table_name>')

Examples

CREATE TABLE mytable(a int, b int) CLUSTER BY(a+1); INSERT INTO mytable VALUES(1,1),(3,3); INSERT INTO mytable VALUES(2,2),(5,5); INSERT INTO mytable VALUES(4,4); SELECT * FROM CLUSTERING_INFORMATION('default','mytable')\G *************************** 1. row *************************** cluster_key: ((a + 1)) total_block_count: 3 constant_block_count: 1 unclustered_block_count: 0 average_overlaps: 1.3333 average_depth: 2.0 block_depth_histogram: {"00002":3}
ParameterDescription
cluster_keyThe defined cluster key.
total_block_countThe current count of blocks.
constant_block_countThe count of blocks where min/max values are equal, meaning each block contains only one (group of) cluster_key value.
unclustered_block_countThe count of blocks that have not yet been clustered.
average_overlapsThe average ratio of overlapping blocks within a given range.
average_depthThe average depth of overlapping partitions for the cluster key.
block_depth_histogramThe number of partitions at each depth level. A higher concentration of partitions at lower depths indicates more effective table clustering.

Was this page helpful?