At 12:37 AM 6/25/2009, you wrote:
Actually, my characterization of the current state is wrong. It appears
that one core is completely busy, I suppose MySQL does this indexing work
in a single thread. Is it reasonable for indexing to be CPU bound?
my.cnf based on my-huge.cnf, expanding key_buf
Actually, my characterization of the current state is wrong. It appears
that one core is completely busy, I suppose MySQL does this indexing work
in a single thread. Is it reasonable for indexing to be CPU bound?
Thanks,
Mike Spreitzer
Mike Spreitzer/Watson/i...@ibmus
06/25/09 01:30 AM
T
Using MyISAM on a table loaded from 8GB of CSV, I am now adding some
indices. In a separate shell I monitor the progress, alternately with
`vmstat` and "show full processlist". At first vmstat shows rapid
progress; an example is
# vmstat 5
procs ---memory-- ---swap-- -io--
We have a composite primary key consisting of
column a, column b, column c. We don't have a lot of variation on
column a and it makes sense for us to cluster by a.
Our queries are
SELECT column c FROM table WHERE column a=something and column e=something.
By creating a composite secondary index on
Hello
I want to fetch and calculate the data from remote database(for example
Japan) and insert them into my local database(for example usa), does
mysql have such function to do it, or do I have to write a script using
perl or other language to help achieving it?
thank you in advance
nathan
Hey Tim, all
On Wed, Jun 24, 2009 at 10:03 AM, Little,
Timothy wrote:
> Why, you might ask, index on physmessage_id? Because then the db won't
> have to do a fetch on items from the table since it's in the INDEX
> itself, saving any unnecessary reads.
FYI: That only holds true for InnoDB, not for
To answer your questions in no particular order, YES you can speed it up
with indexing.
You might want to first create an index on ( blocksize AND
physmessage_id ).
Why, you might ask, index on physmessage_id? Because then the db won't
have to do a fetch on items from the table since it's in the
Hi, all. I'm a long-time MySQL user who's only recently had to start
learning some administrative stuff, largely because I finally have a
decently-sized database. My database is about 100 GB; I'm using it -- via
dbmail (www.dbmail.org) -- as a mail server for my company. While dbmail
is well-and
hi all:
I have three mysql database,tow run as master and the other one runs as
slave.
Some tables in the database have an autoincreament field named as 'rowid'.
These
tables have 100 records on master, but some of these tables on the
slave only have thousands
of record. The tables on the s
