How to speed up count(*) on group by ?

From: Michael Keppler <Michael.Keppler_at_bigfoot.com>
Date: Mon, 11 Oct 1999 10:51:13 +0200
Message-ID: <7ts8d4$ml0@TGZ3>

Hello everybody !

I have a table with about 100.000 lines and one of its columns can have 12 different values (integers). I have to group by this column very often and to count the number of lines for each groups, using something like

select count(*),value
from table
group by value

I defined an index on that column, but it still needs about two seconds to calculate (I think just for numbering the lines in the groups) on a Pentium 333. Is there another way to speed up this kind of selection ?

Ciao and Thanks, Michael.

--
Michael Keppler, MCSE, MCP+I
IT logic GbR
Michael.Keppler_at_gmx.de Received on Mon Oct 11 1999 - 03:51:13 CDT