Hi: Issuing a simple group by like this:
select C_SF, count(*), sum(je) as sum_je
from kp_data
group by C_SF;
against a large (1.4G) table holding a 5 mln records with 60 columns
takes about 330 secs on my Win2000 development box,
a 2.0GHz P4 w/ 1G RAM and an IDE MAXTOR drive.
Reducing the column count helped cut time down to 20 secs, but
that is not exactly what we need for this OLAP web app.
I tried the following optimization hints from the ref manual
with moderate results:
* adding 'order by null' to avoid the final filesort pass.
* increasing buffer sizes to support in-memory operations
key_buffer=64M
table_cache=64
sort_buffer=64M
read_buffer_size=16M
How have you been optimizing your queries and DB setups in
comparable situations?
TIA
Jan Torres
--
MySQL General Mailing List
For list archives: http://lists.mysql.com/mysql
To unsubscribe: http://lists.mysql.com/[EMAIL PROTECTED]
