Elad . wrote:
Hi,
I have a database with 52 columns and 7 million rows.
dear Elad,
Thanks for using MonetDB. The problem you describe is known. No easy solution exists, because the front-end optimizer does not use, nor rely on table statistics to derive the best plan. A solution will emerge, but don't hold your breath
regards, Martin
The columns are Date (timestamp data type), Customer (real data type) and a bunch of other columns that indicate values for each customer on each date (all of these columns are of real data type).
The 'Date' column contains 5 unique values.
The 'Customer' column contains 1.7 million unique values.
The following query returns instantly: SELECT "CUSTOMER" FROM "TABLE".
The following query takes a long time (stopped after a few
minutes): SELECT "CUSTOMER" FROM "TABLE" GROUP BY "CUSTOMER".
- The following query returns instantly: SELECT "DATE", "CUSTOMER"
FROM "TABLE" GROUP BY "DATE", "CUSTOMER".
- The following query takes a long time (stopped after a few
minutes): SELECT "CUSTOMER", "DATE" from "TABLE" GROUP BY "CUSTOMER", "DATE".
- Note that the only difference between query 1 and query 2 are the
addition of a group by clause to query 2.
- Note that the only difference between query 3 and query 4 are the
order of columns in the select and group by clauses.
Is there a way to work around this? I'm really stuck.
Thanks in advance,
Elad
Talk to your Yahoo! Friends via Windows Live Messenger. Find Out How http://www.windowslive.com/explore/messenger?ocid=TXT_TAGLM_WL_messenger_yahoo_082008
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK & win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100&url=/
Monetdb-developers mailing list Monetdb-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-developers