In article <471E26E4.7040804@xxxxxxx>,
Ron St-Pierre <ron.pgsql@xxxxxxx> writes:
>> For what it's worth NUMERIC columns take more space than you might expect.
>> Figure a minimum of 12 bytes your rows are at about 1.5k each even if the
>> non-numeric columns aren't large themselves. What are the other columns?
> The NUMERIC columns hold currency related values, with values ranging
> from a few cents to the billions, as well as a few negative numbers.
What's the required precision? If it's just cents (or maybe tenths
thereof), you could use BIGINT to store the amount in this precision.
This would give you exact values with much less space.
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
