Re: innodb performance issues

2005-07-16 Thread tony
On Fri, 2005-07-15 at 13:28 -0700, David Griffiths wrote: > David, Thanks for your suggestions, i'll give them a try. > There are other tuning choices (including the thread-pool-cache). The > best resource is the page on innodb performance tuning, and it can be > found here: > > http://dev

Re: innodb performance issues

2005-07-15 Thread David Griffiths
Tony, Your my.cnf file is set up for MyISAM, not InnoDB. MySQL allows you to allocate memory and resources to any and all storage engines. Yours is set up to give lots of resources to MyISAM, and none to InnoDB. Reducing MyISAM key_buffer = 384M - this is way too much - I'd set to to 2-16 meg

Re: innodb performance issues

2005-07-15 Thread tony
Hi David, On Fri, 2005-07-15 at 10:25 -0700, David Griffiths wrote: > Tony, > > " - not sure what version you are using 4.1.11. Server is a duel xeon machine with 4gb or ram running mysql and apache webserver and not much else. You should have used > my-innodb-heavy-4G.cnf as the starting p

Re: innodb performance issues

2005-07-15 Thread David Griffiths
Tony, You said that you copied the my.cnf file from "huge.cnf" - not sure what version you are using (I missed your original post), but the my-huge.cnf in mysql 4.0.24 is for MyISAM. You should have used my-innodb-heavy-4G.cnf as the starting point for an InnoDB system. The my-huge.cnf alloc

Re: Innodb Performance Measurement

2005-06-27 Thread Gleb Paharenko
Hello. A lot of statistics you could get from 'SHOW INNODB STATUS'. For example 'FILE I/O', 'INSERT BUFFER AND ADAPTIVE HASH INDEX', 'BUFFER POOL AND MEMORY' could be helpful. See: http://dev.mysql.com/doc/mysql/en/innodb-monitor.html Manoj <[EMAIL PROTECTED]> wrote: > Greetin

Re: InnoDB Performance

2005-04-19 Thread Jigal van Hemert
From: "David Lloyd" > journalling file system. It's not always that clear cut. I've just > switched a number of big customer databases to InnoDB and noone's > noticed any difference - if anything it's going faster. For small tables (<50,000 records) MyISAM is usually a lot faster. However, MyISAM

Re: InnoDB Performance

2005-04-18 Thread David Lloyd
Eko and all, > >The MyIsam storage engine is a non transactional engine and InnoDb is > >a transactional engine. That is the main difference. So I think the > >MyIsam engine should be faster. However, some file systems that have journals are faster than non journalling file system. It's not alw

Re: InnoDB Performance

2005-04-18 Thread Reto Breitenmoser
see this link http://dev.mysql.com/doc/mysql/en/ansi-diff-transactions.html Reto Eko Budiharto wrote: The MyIsam storage engine is a non transactional engine and InnoDb is a transactional engine. That is the main difference. So I think the MyIsam engine should be faster. what is transactional mea

Re: InnoDB Performance

2005-04-18 Thread Eko Budiharto
>The MyIsam storage engine is a non transactional engine and InnoDb is a >transactional engine. That is the main difference. So I think the MyIsam >engine should be faster. what is transactional mean? - Do you Yahoo!? Yahoo! Small Business - T

Re: InnoDB Performance

2005-04-13 Thread Gleb Paharenko
Hello. Send the piece of 'SHOW PROCESSLIST', 'SHOW STATUS' output and corresponding configuration file (after applying all previous advices). It could provide more information to reflection. Marcin Lewandowski <[EMAIL PROTECTED]> wrote: > Hi, > > I've got webserver. There, I've got p

Re: InnoDB Performance

2005-04-13 Thread Marcin Lewandowski
Jocelyn Fournier napisał(a): Hi, What about using another forum ? phpbb2 is well known to be far for what could be called "optimized" :) I hate phpbb, but currently we can't change it :( -- Marcin Lewandowski [ mailto:[EMAIL PROTECTED] gg:188068 jid:[EMAIL PROTECTED] ] [ http://www.nosoftwarep

Re: InnoDB Performance

2005-04-12 Thread Heikki Tuuri
kup tool for InnoDB which also backs up MyISAM tables http://www.innodb.com/order.php Order MySQL Network from http://www.mysql.com/network/ - Original Message - From: "Marcin Lewandowski" <[EMAIL PROTECTED]> Newsgroups: mailing.database.myodbc Sent: Wednesday, Ap

Re: InnoDB Performance

2005-04-12 Thread Daniel Kasak
Marcin Lewandowski wrote: > Hi, > > I've got webserver. There, I've got phpbb2 with circa 6000 users > (average 70-100 users online). There was problems with locking or > something else, when phpbb was using myisam tables. Yesterday, we have > converted tables to innodb, because it should be more

Re: InnoDB Performance

2005-04-12 Thread Gary Richardson
> I've got IDE hdd. Is there simple way to check, if it's overloaded? > Would RAID1 help? (I don't know if in raid1 there are parralel reads or > maybe one disk is only a mirror) If it's IDE, probably not. Moving the database to a different subsystem would give more IO. You can use iostat to check

Re: InnoDB Performance

2005-04-12 Thread Marcin Lewandowski
Gary Richardson napisał(a): Hey, How much load is system vs user? I found that when my company converted some large tables on our old server, the concurrent disk IO increased. Your database server is doing more in parallel and accessing more from your disk at one time. That would be my guess. One o

Re: InnoDB Performance

2005-04-12 Thread Marcin Lewandowski
I've changed settings to: innodb_data_file_path = ibdata1:128M:autoextend innodb_buffer_pool_size=150M innodb_additional_mem_pool_size = 50M and system load is "only" 2 to 3. kernel napisał(a): What does the cpu % show when the machine has the high load avg ? Now, there are about 50% of normal lo

Re: InnoDB Performance

2005-04-12 Thread kernel
Marcin Lewandowski wrote: Hi, I've got webserver. There, I've got phpbb2 with circa 6000 users (average 70-100 users online). There was problems with locking or something else, when phpbb was using myisam tables. Yesterday, we have converted tables to innodb, because it should be more effective.

Re: InnoDB Performance

2005-04-12 Thread Reto Breitenmoser
The MyIsam storage engine is a non transactional engine and InnoDb is a transactional engine. That is the main difference. So I think the MyIsam engine should be faster. Try to adjust the "innodb_thread_concurrency" parameter when you have a lot of users. Reto Marcin Lewandowski wrote: Hi, I'v

Re: InnoDB Performance issues

2003-07-13 Thread Heikki Tuuri
Nicholas, - Original Message - From: ""Nicholas Elliott"" <[EMAIL PROTECTED]> Newsgroups: mailing.database.mysql Sent: Friday, July 11, 2003 6:04 PM Subject: InnoDB Performance issues > --=_NextPart_000_003B_01C3479C.77A1AB60 > Content-Type: text/plain; > charset="iso-8859-1" > Cont

Re: InnoDB Performance issues

2003-07-11 Thread Nils Valentin
Hi Nicholas, How about storing the BLOBS outside of the DB and refering to them ? Best regards Nils Valentin Tokyo/Japan 2003年 7月 12日 土曜日 00:06、Nicholas Elliott さんは書きました: > Hey all, > > I've been experimenting with the best way to store a large (~100GB) of data > for retrieval. Essentially, I'

Re: InnoDB Performance issues

2003-07-11 Thread Nicholas Elliott
ED]> To: "Nicholas Elliott" <[EMAIL PROTECTED]> Cc: <[EMAIL PROTECTED]> Sent: Friday, July 11, 2003 11:29 AM Subject: Re: InnoDB Performance issues > In the last episode (Jul 11), Nicholas Elliott said: > > I've been experimenting with the best way to store a l

Re: InnoDB Performance issues

2003-07-11 Thread Dan Nelson
In the last episode (Jul 11), Nicholas Elliott said: > I've been experimenting with the best way to store a large (~100GB) > of data for retrieval. Essentially, I'm storing 9 variables for > approximately 1,000,000 locations a day for the last ten years. This > can work out at around 4MB a variable

Re: InnoDB Performance Question.

2002-10-08 Thread Heikki Tuuri
Jungshu, - Original Message - From: ""Heo, Jungsu"" <[EMAIL PROTECTED]> Newsgroups: mailing.database.mysql Sent: Monday, October 07, 2002 5:49 AM Subject: InnoDB Performance Question. > Hello everyone. > > I'm working on migrating Oracle to MySQL 4.0.3 > > MySQL works on Redhat 7.3 and

Re: Innodb performance (again)

2001-08-17 Thread Heikki Tuuri
Richard, I calculate 430 s / 116 s = 3.7, not 7 :). Last time you had extremely slow execution probably because of fsync. Obviously the doublewrite method has fixed it, since the performance is not that bad any more. Linux kernel 2.4 might be better. Remember that MyISAM uses the file cache of