MySQL - Extremely inefficient and unreliable query joining 3 tables - how to improve? - mysql

Question

I am using MySQL with an INNODB engine on a DigitalOcean machine. The machine has 4GB memory, an 80 GB DISK and 2vCPUs and runs on Ubuntu 16.04.
We have a query that joins three tables that runs very slowly (takes about 5 minutes to return, if it works at all). The size of the tables are 6 million, 20 million and 100 thousand rows, respectively, and there are unique indexes in the tables for each row.
The query looks like this:
SELECT *, table2.column1
FROM table1
INNER JOIN table2 on table1.column1 = table2.column1
INNER JOIN table3 on table1.column2 = table3.column1
WHERE table3.column2 == "{ID}";
We want to embed this query in a data processing / analysis pipeline which dynamically pulls relevant data into memory and then runs further analysis using R. For this purpose, we need to make this query [or an alternative, which does the same thing] run way more efficiently.
Does anyone have any ideas as to how to make this query more efficient, or what the reasons for this slowdown may be? Any help would be greatly appreciated,
Many thanks!

Answer 1

For this query:
select table1.*, table2.column1
from table1 inner join
table2
on table1.column1 = table2.column1 inner join
table3
on table1.column2 = table3.column1
where table3.column2 = "{ID}";
You want indexes on:
table3(column2, column1)
table1(column2, column1)
table2(column1)