How to write a very large JOIN or equivalent?

Question

PostgreSQL

What is the point.
There is a bunch (m, somewhere around 50) of the same type of plates with two columns: id -- primary key and value -- a certain number of double precision. Each table has the same number of the same id (and there can be quite a lot of lines, several million).
The task seems to be not difficult: to combine all small tables into one large one of the form (id, table_1.value, ..., table_n.value).
There are two solutions. The first is CREATE TABLE blabla as SELECT with a big long JOIN, which seems to work fast enough, but only on Postgres and only on a small dataset (and there is not enough RAM on a large database). On another database engine, Greenplum, JOIN on a large number of columns simply hangs the base.
The second solution is to first make the table a selection of all primary keys, and then sequentially add the corresponding columns. This method takes too long.
Is there any quick solution to the problem?

Answer 1

Wrong data structure, so they stepped on a rake.

Answer 2

Write an external script that will run in parallel through 50 tables in ascending order of id, combine rows and pour them into a pivot table?