Hi!
Here is a worst case i have even seem so far:
create table aa(
some_number int unsigned not null,
index aa_sn (some_number)
);
create table bb(
min_number int unsigned not null,
max_number int unsigned not null,
range_type char(1) not null,
primary key (min_number, max_number)
);
now we put 25000 records into aa and 1000 into
bb and try to figure out in what range each number in aa:
select some_number, range_type from aa inner join bb on
some_number>=min_number
and some_number<=max_number;
I use inner join because number which are not in one of the ranges are
not interesting to me.
Explain gives 25000x1000 rows to scan and range check w/o
using index. So, the select take about 7 minutes to complete.
Is there any way to make such case faster. The problems is that
I don't know ranges until aa has all the data, so i cannot
figure out the range while inserting into aa. Tricky business.
Any ideas?
Not really impirtant, but i thought i would an interesting task.
Regards,
Artem
---------------------------------------------------------------------
Before posting, please check:
http://www.mysql.com/manual.php (the manual)
http://lists.mysql.com/ (the list archive)
To request this thread, e-mail <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
