What to do, and not join SQL when scaling horizontally?

Question

What will I use instead of SQL connections when I have a large complex relational database that is simply too large to be installed on one machine? I began to outline the database on many machines, but as a result, I can no longer connect effectively.

Any tips?

Answer 1

There are many approaches to this work, the general idea is to outline your data in such a way as to group related data.

() , Game, Player PlayerGame (playerId). , , " " . . , .

, , , , .

:

• ( -, ).
• .

- , (, ).

, :

http://dbshards.com/dbshards/database-sharding-white-paper/

, . , - , . ( ) ​​ . , , GameInstance. " " GameInstance PlayerGame , GameInstanceId. .

, , , , "" .

Answer 2

, , . , .