I have a table with a full-text index that spans four text columns. It
seems that the more records a query matches, the longer it takes to
execute. Is that normal?
Examples:
I have 15,376 records in this table:
mysql select count(*) from NS_articles where
Hi,
How do I get a full-text relevance ranking when using IN BOOLEAN MODE?
The manual shows how to get the relevance, but if I try to apply the
same technique to a boolean search, I get a score of either 0 or 1 as
the relevance. There are relevance parameters for boolean searches,
so it must be
Hi,
I have a question about indexing a column of type SET. It seems that
generally the index is not too useful, since it always requires some
sort of expression to get what you want out of the set, and therefore
the index won't be used.
Example:
create table FOO (BAR
Hi,
I have hundreds of mysql databases on a server, mostly filled with
newspaper articles for web sites.
For one particular database, I have about 10,000 archived articles I
want to load in. I want to do this without affecting the performance of
the live site (or any other sites using this
be a # of connections issue. And
why is the entire server affected, not just that table?
TIA
Monte
Gerald Clark wrote:
You need to throttle back the inserts.
You are doing 5-10 inserts per second, but
that is fully loading the server.
Do 1 or 2 inserts, and sleep a second.
Monte Ohrt wrote
a chance to get in.
Monte Ohrt wrote:
Thanks for the quick reply Gerald.
What specifically is being loaded? The system load certainly isn't the
problem. Is it a buffer that fills, or a lock queue, or something
else? Although the inserts are done in rapid succession, there is only
one
Hi, I have a problem where the index for a SET column does not get used
once I link in another table.
The output to explain:
mysql explain select count(*) from NS_articles where
find_in_set('approved', artFlags) 0;
\
--with-mysqld-user=mysql\
--with-libwrap \
--enable-thread-safe-client \
--enable-large-files\
--with-berkely-db
Monte
--
Monte Ohrt [EMAIL PROTECTED