I have a table which I want to use to read data. Usual query:
SELECT id FROM table1 WHERE (col1 BETWEEN A AND B) AND (col2 BETWEEN C AND D)
SELECT * FROM table2 WHERE id IN()
Average number of returned rows 10^5. DB runs just on cluster of rather old PCs.
Is it correct to use as high replication as possible (optimally equal to number of nodes and perform reads with consistency level ONE?