| |
| |
| |
|
Page: 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Comments:
<0> so yeah, 5.0 fastest is convert, then datediff, then floor
<1> it turned out to be because i was running a recursive stored procedure at the same time :P
<0> 5.1 fastest is datediff by a big margin, then convert, then floor
<2> hmm.. what are the commands to test?
<0> select benchmark(1000000,datediff('2006-05-31 12:14:16','1998-08-21 23:02:47'));
<0> select benchmark(1000000,convert((1149095656-903758567)/86400,unsigned));
<0> select benchmark(1000000,floor((1149095656-903758567)/86400));
<0> i should go pick up my wife who just called
<0> bbiab
<2> http://hashmysql.org/paste/viewentry.php?id=1905
<1> ok so that is 5 :P
<1> what are the results from 4 like ?
<2> here:http://hashmysql.org/paste/viewentry.php?id=1906
<1> so convert and floor are much faster on v4 :P
<2> yeah, much
<1> even datediff is substantially faster
<2> let me confirm i didn't have any odd trafic
<1> 30% is nothing to be sneezed at
<2> if i did have any traffic it would be on the 4.1 machine but this is the same box just different server
<2> the command: elect benchmark(1000000,floor((1149095656-903758567)/86400)); runs max: .26 min .12
<2> but usually .14 - .19
<1> does not mysql remember queries when you run exactly the same one again and not have to do so much optimisation on subsequent runs ?
<1> shouldn't there be some kind of query compilation cache ?
<1> although i guess for a benchmark it would not make a huge difference
<1> since you are running the query so many times :P
<2> perhaps, but it is seems to be consitent to the first query avereraging the original number
<2> i got one .12, all others are .19 -.16
<2> and a .26 one time
<1> is this a test server or one in use by others ?
<2> the lowest is .12
<2> it is being used
<1> that makes more sense
<1> on my box it takes exactly the same amount of time each time
<2> now I'm hitting .12 more consitantly
<2> let me hit the 5.x server a few times and see what that does
<2> there is no doubt.. 5 is slower with this
<1> heh
<2> i can't get lower than .95
<2> err. .85
<1> the mysql staff are probably watching us right now and cringing that someone actually noticed
<2> there is the issue that my mysql4 is using as much memory as possible, i dont know if that would effect this query
<1> it should not affect it too much
<2> mysql5 is using default
<1> i doubt that that particular query requires much in the way of memory
<1> it is not like it is doing anything :P
<2> still havn't gotten below .95
<2> 8*
<1> it is strange though - when i upgraded the celeron 1200 to v5 i noticed an improvement in performance for some queries
<1> and it previously had 4.1
<1> of course i also migrated it from redhat to slackware with a custom 2.6 kernel at the same time
<1> which could be part of the reason :P
<2> yeah that could explain a lot
<1> the whole box just seems snappier
<1> also newer versions of apache and perl
<1> and i have several perl jobs that run continuously on that one
<1> must rewrite them in c++ one day :P
<3> is it possible to have mysqld listening on multiple (but not all) IP addresses?
<3> or do i have to iptables -t nat around it?
<0> man i wonder why floor/convert **** so much on 5.0 vs 4.1
<0> it's such a weird thing
<1> indeed
<0> floor seems like such a basic thing
<4> 2006-05-26:43:55Torrent 7954 (Armin Van Buuren-Live At The Asta %28ASOT 250%29-FM-26-05-2006-1KING) was deleted by Frawg (TT Release now out)
<4> 2006-05-26:43:42Torrent 7953 (John Askew-Live At The Asta %28ASOT 250%29-FM-25-05-2006-1KING) was deleted by Frawg (TT Release now out)
<4> 2006-05-26:43:26Torrent 7952 (M.I.K.E-Live At The Asta %28ASOT 250%29-FM-25-05-2006-1KING) was deleted by Frawg (TT Release now out)
<4> 2006-05-26:43:16Torrent 7951 (Jonas Steur-Live At The Asta %28ASOT 250%29-FM-25-05-2006-1KING) was deleted by Frawg (TT Release now out)
<4> 2006-05-26:42:09Torrent 7948 (Armin Van Buuren-Cl***ic Hour Live At The Asta (ASOT 250)-FM-25-05-2006-1KING) was deleted by Frawg (TT Release now out)
<0> eh?
<0> stop that.
<4> sorry, wrong window :)
<5> is there such thing as a mysql "isin"
<5> or need to do those sort of comparisons in php?
<3> there is the keyword IN
<1> that depends on what you want it to do
<1> as a general rule you can do pretty much any comparison in sql
<5> I wanna do SELECT * FROM table_name WHERE $query isin $db_info
<6> FatalRemedy: There are many forms of that directly in SQL.
<7> can someone help me out here, i am following this tutorial and i am on the 3rd command on the page, and i get an error saying 'lost connection to MySQL server during query'
<7> this is the link : http://www.howtoforge.com/perfect_setup_debian_sarge_p4
<7> i dont know whats wrong, i followed this tutorial step by step.
<7> the command i type in is this
<7> mysqladmin -h server1.example.com -u root p***word yourrootsqlp***word
<8> myusa___: replace server1.example.com with your hostname.
<7> that is my hostname
<7> is there a command to check your hostname? maybe i spelt it wrong
<8> myusa___: hostname -f
<9> how come I keep getting an ERROR 1064 (42000) when I do CREATE DATABASE ''blog'';
<7> bluestrain__ nope it was right, my hostname is server1.example.com
<9> anyone? please?
<6> blank_: Get rid of the quotes.
<9> Xgc, ooooooooo!!! thanks buddy!
<0> ....
<6> blank_: Then tell me if you still have a problem.
<9> yay it works Xgc, thanks! :)
<6> You're welcome.
<7> can anyone help me out?
<10> hi all
<10> is there a way to drop all table with a prefix
<10> i have twenty tables with the prefix test_
<10> it's possible to call drop table in in this way, drop table test_*
<0> tell kioto about tias
<6> kioto: Place them in a database called test and then drop database test;
<0> err
<0> !tell kioto about tias
<0> :P
<11> Alrighty.
<6> kioto: or use INFORMATION_SCHEMA and dynamically generate the drop statements.
<10> drop db is simple but i have others tables that i need to work
<10> i'll try with drop user gleam_ :P
<6> kioto: What I meant by that is, you can use multiple databases, one for each table_prefix you wish to use.
<10> yes but i have limit on my provider about the number of databases that i can create
<10> thanks Xgc, gleam_
<0> pfft id idnt' help
<0> i didn't
<0> :P
<12> I have 2 tables like table1 (id,name) table2 (id,value,table1_id). then I want select table1's list and table2's value summary (like table1's name1 | table1's summary1, table1's name2 | table1's summary2 etc).
<12> is it possible?
<13> Hello; mysqld isn't running for me. I'm getting "Fatal error: mysql.user table is damaged or in unsupported 3.20 format" in my logs. I found a suggested solution here: http://tinyurl.com/pjsht but running mysqld_safe is merely exiting right away, so I can't seem to apply the fix.
<13> Issuing "mysqld_safe -skip-grant-tables -user=root &" responds: "Starting mysqld daemon with databases from /var/lib/mysql" then "STOPPING server from pid file /var/run/mysqld/mysqld.pid" and "060527 03:19:17 mysqld ended".
<4> --
<4> --
<4> :)
<4> *ponders*
<14> tulga: rephrase your question
<15> petruchio: let me guess, you just upgraded.
<13> Most likely. I didn't, but I think it happened with a general update of a Gentoo system. I'm not the only admin.
<15> !m petruchio upgrading 4.0 4.1
<11> petruchio: (Upgrading from Version 4.0 to 4.1) : http://dev.mysql.com/doc/mysql/en/Upgrading-from-4.0.html
<13> Thanks.
<15> read the 2nd bullet point after the first "Incompatible Change"
<15> (and preferably the whole page)
<15> There is also "myisamchk", a command-line tool to repair broken myisam tables. However, it will likely continue to do this if you don't follow the instructions on that page.
<13> Okay. I don't *think* we were doing anything with multi-byte character sets, but it's possible. I'll read the page. Thanks again.
<13> Ah, wrong bullet point. :-)
<16> what should i set so that mysql will support sjis?
<16> default-character-set=sjis, seems to be not the solution, it will still display question mark
<15> display how?
<17> hello all, I have a question, does mysql4 support such subqueries as 'update (select)' or 'update (insert)' I need a query which update row and if there are no such row - query creates it)
<18> 4.1 supports subqueries ... but not on the same table that is being updated
<17> I have 4.0.26 :(
<18> use a join
<1> aibo: why not just use the appropriate syntax on your insert statement ?
<17> and I need to solve my prob using 1 query :(
<1> is there not an "on duplicate update" option or something similar ?
<1> am sure i read that somewhere :P
<18> andrewbryson: yes .. insert on duplicate key update
<17> andrewbryson, link?
<18> !m aibo insert
<11> aibo: (INSERT Syntax) : http://dev.mysql.com/doc/mysql/en/INSERT.html
<1> it is in the manual...
<17> thx
Return to
#mysql
or
Go to some related
logs:
no acceptable C compiler foudn in $PATH
fglrx forcemonitor
#css
imdct gentoo
ubuntu pctv 300e
#debian
#suse
#web
#linux
#physics
|
|