Skip to main content

VACUUM DROP TABLE

Introduced or updated: v1.2.368
ENTERPRISE EDITION FEATURE
VACUUM DROP TABLE is an Enterprise Edition feature. Contact Databend Support for a license.

The VACUUM DROP TABLE command helps save storage space by permanently removing data files of dropped tables, freeing up storage space, and enabling you to manage the process efficiently. It offers optional parameters to target specific databases, preview, and limit the number of data files to be vacuumed. To list the dropped tables of a database, use SHOW DROP TABLES.

See also: VACUUM TABLE

Syntax

VACUUM DROP TABLE 
[ FROM <database_name> ]
[ DRY RUN [SUMMARY] ]
[ LIMIT <file_count> ]
  • FROM <database_name>: This parameter restricts the search for dropped tables to a specific database. If not specified, the command will scan all databases, including those that have been dropped.

    Example:
    -- Remove dropped tables from the "default" database
    VACUUM DROP TABLE FROM default;

    -- Remove dropped tables from all databases
    VACUUM DROP TABLE;
  • DRY RUN [SUMMARY]: When this parameter is specified, data files will not be removed; instead, it returns a result showing which data files would have been removed if this parameter were not specified. See examples in the Output section.

  • LIMIT <file_count>: This parameter can be used with or without the DRY RUN parameter. When used with DRY RUN, it limits the number of data files to be shown in the DRY RUN result. When used without DRY RUN, it limits the number of data files to be vacuumed.

Output

The VACUUM DROP TABLE command returns a result when the DRY RUN or DRY RUN SUMMARY parameter is specified:

  • DRY RUN: Returns a list of up to 1,000 candidate files and their sizes in bytes for each dropped table.
  • DRY RUN SUMMARY: Returns the total number of files to be removed and their combined size for each dropped table.
Example:
VACUUM DROP TABLE DRY RUN;

┌──────────────────────────────────────────────────────────────────┐
tablefile │ file_size │
├────────┼─────────────────────────────────────────────┼───────────┤
│ b │ 313ebd4da5cc493f9a7d491da8253ce2_v2.parquet │ 210
│ b │ 737f2215b8ac4a268d5b7f2218273358_v2.parquet │ 210
│ b │ 737f2215b8ac4a268d5b7f2218273358_v4.parquet │ 340
│ b │ 313ebd4da5cc493f9a7d491da8253ce2_v4.parquet │ 340
│ b │ last_snapshot_location_hint │ 72
│ b │ 7e01fa5c2e0a495298942671447dc8cb_v4.mpk │ 515
│ b │ 2bc90e5be55c44258a736d27e5f7ac9e_v4.mpk │ 459
│ b │ 85e73803aabc4eb48774db3d932312dd_v4.mpk │ 534
│ b │ f0e507d0b825428dbfe57c8d8b620a15_v4.mpk │ 533
│ c │ cee790e76f6e4e92bc9dab3b9e873dcf_v2.parquet │ 210
│ c │ 4bcb2cef3b6344cb951908ebee5ceb36_v2.parquet │ 210
│ c │ cee790e76f6e4e92bc9dab3b9e873dcf_v4.parquet │ 340
│ c │ 4bcb2cef3b6344cb951908ebee5ceb36_v4.parquet │ 340
│ c │ last_snapshot_location_hint │ 71
│ c │ 414fc6a8dc6746afbc576cf8fddfcdf3_v4.mpk │ 516
│ c │ 8d0d115c438244c295e3bfd50d556e39_v4.mpk │ 458
│ c │ 28e4f551cc634d3d8d7e648c3baa5f5c_v4.mpk │ 534
│ c │ 007b57e08eda419fbb451a3a3ed71de8_v4.mpk │ 533
└──────────────────────────────────────────────────────────────────┘
VACUUM DROP TABLE DRY RUN SUMMARY;

┌───────────────────────────────────┐
table │ total_files │ total_size │
├────────┼─────────────┼────────────┤
│ b │ 93213
│ c │ 93212
└───────────────────────────────────┘

Adjusting Data Retention Time

The VACUUM DROP TABLE command removes data files older than the DATA_RETENTION_TIME_IN_DAYS setting. This retention period can be adjusted as needed, for example, to 2 days:

SET GLOBAL DATA_RETENTION_TIME_IN_DAYS = 2;

DATA_RETENTION_TIME_IN_DAYS defaults to 1 day (24 hours), and the maximum value varies across Databend editions:

EditionDefault RetentionMax. Retention
Databend Community & Enterprise Editions1 day (24 hours)90 days
Databend Cloud (Standard)1 day (24 hours)1 day (24 hours)
Databend Cloud (Business)1 day (24 hours)90 days

To check the current value of DATA_RETENTION_TIME_IN_DAYS:

SHOW SETTINGS LIKE 'DATA_RETENTION_TIME_IN_DAYS';
Did this page help you?
Yes
No
Explore Databend Cloud for FREE
Low-cost
Fast Analytics
Easy Data Ingestion
Elastic Scaling
Try it today