I would like to practice a little on cats in the area of ​​compiling sql queries, indexes and all that. For this, I would like to find some large database with a fairly large number of records in the tables (more than 100 million). Maybe someone knows if there are such backups in the public domain?

  • @ru_volt thanks, I'll try - Pupkin
  • Delivered your comments in response. The so-base has a structure, and even it is updated sometimes, so I would try it - Duck Learns to Take Cover

3 answers 3

Stackexchange (of which this site is a part) publishes a dump of its own database , although I myself have not tried to play with it. The structure of the database in this dump is described in general here.
If there are problems with importing this into sql, then for the data-dump label there are several answers on import-related issues.

Another list of databases that can be used for testing may be part of outdated, but mostly useful.

    You can try to take Wikipedia dumps: https://dumps.wikimedia.org/backup-index.html

    Link to the instruction with the description of the tables: https://ru.wikipedia.org/wiki/Wikipedia : How_to make_copy_Wiki.

      The following code quickly fills the fields with unique values. Just suitable for analyzing optimizations, indexes, etc.

      While 1=1 INSERT INTO 'your table' VALUES (NEW ID(), NEW ID(), NEW ID(), DatePart(millisecond,GetDate())