Cloud SQL information schema
MySQL | PostgreSQL | SQL Server
This page lists the Cloud SQL information schema table additions.
information_schema.innodb_vector_indexes
Gives all the vector indexes that are opened in the memory after restart.
| Column name | Description |
| INDEX_NAME | Name of the index |
| TABLE_NAME | Qualified table name in db_name.table_name format |
| INDEX_TYPE | TREE_SQ is supported |
| DIMENSION | Dimensionality of the vector column |
| DIST_MEASURE | Distance Measure on which index is built |
| STATUS | A string describing the current state of the index |
| STATE | Internal state of the index |
| NUM_LEAVES | Number of leaves as configured by the user or computed internally based on the size of the base table |
| NUM_LEAVES_TO_SEARCH | Number of leaves to search in ANN. Can be overridden at query time |
| QUERIES | Number of ANN queries on this index since server start |
| MUTATIONS | Number of DML operations on base table that resulted in updating the vector index |
| TREE_MEMORY | Memory occupied by the non-leaf part of the vector index |
information_schema.innodb_all_vector_indexes
Contains all the vector indexes that exists on the instance (even if they arenot opened in the memory yet).
| Column name | Description |
| id | same as innodb_indexes.index_id |
| table_id | An identifier representing the table associated with the index |
| sub_table_id | An identifier representing the sub table associated with the vector index |
| state | Internal state of the index. Same asinformation_schema.innodb_vector_indexes |
| corrupted | Indicates whether the index is corrupted or not.1 means corrupted,0 means not corrupted. |
| config | A json value showing index configuration. In configuration:
|
information_schema.innodb_vector_indexes_memory
Provides information about overall memory usage of vector indexes in theinstance.
| Column name | Description |
| STATE | Memory management for vector indexes is enabled. |
| TOTAL_MEMORY | cloudsql_vector_max_mem_size |
| INDEX_MEMORY | Amount of memory that is used to loadindex_tree into the memory. |
| TRAINING_MEMORY | Amount of memory that is allocated for training during index build. |
| LOADED_INDEX | Number of indexes that are loaded in the memory. |
What's next
- Read theoverview about vector search on Cloud SQL.
- Learn how toenable and disable vector embeddings on your instance.
- Learn how togenerate vector embeddings.
- Learn how tocreate vector indexes.
- Learn how toperform searches on vector embeddings.
Except as otherwise noted, the content of this page is licensed under theCreative Commons Attribution 4.0 License, and code samples are licensed under theApache 2.0 License. For details, see theGoogle Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates.
Last updated 2025-12-17 UTC.