Tom,
thank you.
> I think the problem is probably that you're sorting two dozen CHAR
> columns, and that in many of the rows all these entries are '' forcing
> the sort code to compare all two dozen columns (not so)? So the sort
> ends up doing lots and lots and lots of CHAR comparisons. Which can
> be slow, especially in non-C locales. What's your locale setting?
show all returns
"lc_collate";"en_US.UTF-8"
"lc_ctype";"en_US.UTF-8"
"lc_messages";"C"
"lc_monetary";"et_EE.utf-8"
"lc_numeric";"et_EE.utf-8"
"lc_time";"et_EE.utf-8"
How to speed up this query ?
Is it possible to force the binary comparison for grouping ?
Should I concatenate all the char columns into single column ?
Andrus.
---------------------------(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
