| |
| |
| |
|
Page: 1 2
Comments:
<0> anyone know of any performance test tools available for mysql? want to ensure its running correctly for the website. The only way i can think of is checking times etc, anyone got any ideas?
<1> SELECT BENCHMARK()
<1> SELECT BENCHMARK(1000000,ENCODE("foo","bar")); Query took 0.7606 sec
<0> so if i do SELECT BENCHMARK(1000000,ENCODE('hello','goodbye')); and it says 1 row in set (0.64 sec)
<0> is that good?
<2> Is there a way to use LIKE to match strings containing only A's for exemple? I dont see how do have a "variable length"
<3> hey I got a simple one that'sjust got me stuck
<3> I wanna update the value of a an int by +1
<3> UPDATE `movies` SET `clicks`=+1 WHERE......
<3> but it just sets it to 1
<3> how do I make it + whatever it's current value is?
<3> s`pl
<3> s`ok... got it :)
<1> Advocated well it was faster than mine
<1> SELECT BENCHMARK( 1000000, ENCODE( "foooooooooooooooo", "barrrrrrrrrrrrrrrrrrrr" ) ) Query took 2.5751 sec
<4> my system is slow..
<4> that exact gives me 1.19 here
<0> 1 row in set (1.64 sec) ^
<4> anyways, ENCODE will only measure CPU
<0> ahh
<4> the real bottleneck for DBMSes tends to be HDD-speed
<0> basically, when im reading a record from a table to show on a page, i want to test how long its going to take, and then try and simulate the same if there were to be 1000 people accessing the same page
<1> thats on a hosted account, so i think its not to bad
<4> Eldritch: yeah, but it isn't "real"
<4> as I said, it will only measure CPU
<1> neither are you, your just a figment of my imagination
<4> hehe
<4> Advocated: ever used httperf?
<4> is this your own server by the way?
<4> running stress-testing on a shared system is a Bad Thing(tm9
<4> s/9/\)/
<4> http://www.hpl.hp.com/research/linux/httperf/
<5> 1.75 secs on a dual opteron
<4> Freman: which one?
<5> but then, it's a stressed dual opteron (C:
<4> hehe
<5> the one with 55 million row tables
<4> it does the job nicely though
<4> it didn't drop below 80% idle when it has 4,000 visitors in an hour
<5> I have 8 threads currently running when I did that (c:
<4> Threads: 56 Questions: 1040446 Slow queries: 0 Opens: 192 Flush tables: 1 Open tables: 256 Queries per second avg: 0.602
<0> K_F never heard of httperf to be honest :s and yea its my own server
<5> anywya
<5> must go to work
<4> look into it, it tends to be nice for stress-testing
<5> I'm only an hour late
<4> hehe
<4> enjoy
<5> meh
<4> I'm on the other end of the day
<4> 0145 here
<4> sipping my beer
<4> not that it matters, I have easter holidays
<4> no lecture untill monday next week (not coming)
<6> What do you think is more expensive one SQL query getting 60 columns, or 60 SQL queries getting 1 column.
<1> ones being hit with a feather the other a brick
<7> hi
<7> how can i rewrite the an autoincrement field - i want to be consecutive numbers - some rows was deleted
<8> hello, how do i set mysql idle connection timeout?
<9> salut
<10> is there a tool that takes my current database and creates sql commands to create it ? like CREATE TABLE blah blah
<11> mysqldump
<10> thanks
<12> Hey folks, I am getting a 1064 syntax error with a Create Table statement.. would it be possible for someone to take a look at my table to see if they can find the syntax error?
<12> er.. my create statement
<13> hello
<14> hey dravine
<14> whats up
<13> anyone have any resources for performance tuning mysql?
<13> howdy LanShark
<14> not i, you got anything for the man grazzy?
<13> ?
<14> grazzy or K_F maybe able to help
<14> i just troll here
<13> ah
<14> argh
<14> IAMNOTHERE
<13> I'm not a mysql person, but I set up mediawiki and imported the wikipedia db into an old laptop of mine, and was kind of blown away by how slow it was
<13> it's a 7.7 gig DB, but that doesn't seem very big to me
<14> what? did you say something?
<13> yeah
<13> I prefer postgres
<13> however, media wiki is the **** with pg support
<13> I added an index to one of the wiki tables in mysql, and it took 17 hours to finish
<14> arrrgh
<13> granted the box is only a 667Mhz, 256Megs of ram, and an 80 gig drive
<13> but still
<13> the sad thing is it didn't seem to have any impact on the performance
<14> heh
<13> and everytime mysql starts up, it chews up 99% of the proc and pegs the disk io at a solid 10megs a second for about half an hour
<14> good thing it took so long!
<13> yeah
<13> that's what I was thinking
<14> my god man
<13> so far I'm not impressed with mysql heh
<13> I'm hoping to find that it's just something in default settings that can be tuned to fix these issues
<13> maybe a --dont-**** compile option or something
<14> heh
<14> --fullspeedahead!
<14> i havent used it for anything that large, but i would think it could hanlde it
<14> can you take the laptop out of the scenario without wasting 17 hours?
<13> well the laptop is the target platform
<13> I'm sure if I put it on my 9300 it would run much better
<13> but man, I can't beleive it ****s so much on older hardware
<13> especially when the only time it gets traffic, is when it's from localhost
<14> maybe create another db and fill it with less data?
<14> see if it still ****s
<13> I was aiming for a good offline resource for college
<13> if I put any effort into reimporting the 4.5G of xml that is wikipedia, I'll be doing it into postgres
<13> because I already know how to tune that
<13> and it seems to have far better indexing facilities than mysql
<14> yea, im not sure what to say, although i wouldnt want to spend too much time tuning mysql when that *may not* be the problem
<14> im a hack though, so take what i say with a grain of salt
<13> well, mysql is the only source of lag on the box
<14> heh
<14> not looking too good for mysql
<13> actually, it tends to lag everything else on the thing
<13> it took 3 days to import the db dump from wikipedia
<14> my god man
<13> well, wikimedia has a ****ty importer
<13> php script that does about 50 selects for every page it inserts
<13> the alternative was to run some c hackery against it, which produces sql dumps
<14> this is where mysql and php really irk me
<13> but they get ****ing huge
<13> they more than match the 4.5 gigs of xml
<13> and the original file stays there
<13> so to do all this import hackfu, I needed at least 15 gigs of disk free
<14> WAFM
<14> m for mess
<14> heh
<13> yeah
<14> you should have a cigar
<13> I don't look forward to porting the stuff to postgres if that's what I end up having to do
<14> you been to freenodes pg channel?
<13> no
<13> I know postgres
<13> I don't know mysql
<13> that's why I was asking here
<14> i hear ya, i recommend that pg channel for anyone doing pg though
<13> not sure why, this channel has little activity most of the time
<14> thats the most helpful channel ive seen on irc
<13> freenode tends to be pretty decent
<15> hello just a quick question ,i have a solid java ,c++ background and i would like to learn php what books would u recommend and how long do u think it will take me to become good??
<15> sorry mysql..
<15> i meant
<13> if I could kickban you in here, I would
<13> just a little fyi
Return to
#mysql
or
Go to some related
logs:
amzuklar
#java
#php
#skype
lolazita
#linux
kernel ensambler
#MissKitten
#c
#AllNiteCafe
|
|