next:
select cols from table where id > YOUROLDID and forums = CURRENT_FORUM order by id
limit 1
previous:
select cols from table where id < YOUROLDID and forums = CURRENT_FORUM order by id
desc limit 1
however, this may not be as fast or efficient enough
ou might want to create another id field
BEGIN my.cnf
[mysqld]
port = 3306
socket= /tmp/mysql.sock
basedir = /usr/local/mysql
log = /var/log/mysql/mysql.log
log-slow-queries = /var/log/mysql/mysql-slow.log
log-err = /var/log/mysql/mysql.err
log-bi
what table types? Innodb.etc...
about how many tables?
do you do alot of sorting?
are the exact same queries repeated alot?
is the machine doing anything else or is mainly a DB server?
can I borrow the machine for awhile?
what version of mysql are you running?
chris
-Original Message-
Fr
>Is it me or do these dual athlons seem rather responsive!
Yay!! I win (so far... heh)!
I ran it on one of our servers (not idle... running apache w/ CGIs and db calls ~
140,000 scripts/day )
We are using as many tricks as we can.
0. mysql 4.0 binary from mysql.com
1. we set the n
1 row in set (0.54 sec)
2x Athlon2100
2GB Ram
Linux 2.4.20
-Original Message-
From: Jake Johnson [mailto:[EMAIL PROTECTED]
Sent: Wednesday, July 23, 2003 9:34 AM
To: Mysql
Subject: What is a good benchmark?
I ran this benchmark on my pIII 500 and was wondering what everyone else
was ge
Thanks for everybody's help!!
This way worked for me. Clever!
Chris
... and yes... even my example was messed up
:-P
-Original Message-
From: Rudy Metzger [mailto:[EMAIL PROTECTED]
Sent: Monday, July 14, 2003 4:15 AM
To: Christopher Knight; MySQL List
Subject: RE: grouping / so
Im having troubles getting the results I want... can someone suggest which way to go..
mysql 4.0
A
| 1 | |
| 1 | BAD |
| 2 | |
| 3 | BAD |
| 3 | |
| 4 | BAD |
| 5 | |
| 5 | BAD |
| 5 | BAD |
what I want is 5 results. If there is a NULL, then ret
/Tips.html
-Original Message-
From: Henrik Leghissa [mailto:[EMAIL PROTECTED]
Sent: Tuesday, June 24, 2003 9:42 AM
To: Christopher Knight
Subject: RE: mysqld question
At 16:39 2003-06-24, Christopher Knight wrote:
>it really depends on
> how big your database is
> what else is ru
it really depends on
how big your database is
what else is running on the server (apache..)
how many connections at once
percentage of selects to updates/inserts
do you have indexes
what version are you running
what is your mom's maiden name
if you look at the sun, how long to you blink
:
OR, if you are good at vi,
you can insert a ' at the begining and end of every line (if you dont have
any 's in the file)
then put a , at the end of every odd line
then join every other line
the put a "insert into blah (question, answer) values ( " at the
begin of every line
and then
I swear we went over this today...
for anybody still unsure...
give a look at the bottom of any email you recieve from the list and it will
say 'To unsubscribe' on the last line... just click the link
It will probably be at the end of this message even..
--
MySQL General Mailing List
For list a
Optimally, Yes, you should replace with exact same brand/model etc...
but you CAN replace with a different brand/ model drive of the same
amount of disk space or more. It isn't recomended (because of
different seek times, cache .. etc..) but if you are carefull and do
your research, you can get a
Im using a 3ware (which has great linux support) Escalade 7800 with 8
120GB/8MB cache ide drives in RAID 10 under Debian with 2.4.20 kernel.
I guess what we need to know is what platform and how much $$ you wanna
spend
-Original Message-
From: Bernd Jagla [mailto:[EMAIL PROTECTED]
Sent: T
hehe... been there...
if there isnt a way if you have logging turned on, you might be able to
find the insert and following updates statements in the log to rebuild the
row.
good luck
chris
-Original Message-
From: Arcangelo [mailto:[EMAIL PROTECTED]
Sent: Thursday, June 12, 2003 10:2
Can you sum on counts?
What I want is the # of objects with a freq of 1, freq of 2, freq of 3...
and then total # of objects.
>>select object_id from object_hist where type_id=5879;
+---+
| object_id |
+---+
| 2121 |
| 3234 |
| 2121 |
| 4876 |
| 4876 |
|
Im have a problem / misunderstanding with the 4.1 release. Im having issues
with indexes related to 'IN' vs '=' in subselects.
assume the subselect returns 1 value
select SQL_NO_CACHE user_id from table_a where user_id in (select
SQL_NO_CACHE user_id from table_b where ... blah)
9.5 seconds...
s
16 matches
Mail list logo
