Hi all!
I've got a few questions that I was hoping some of the fine readers of
this list could help me out with. I'll probably be going into a
development meeting this coming week and will need to have some
information up my sleave to ensure the mighty MySQL is selected as the
database backend for
Hi again all,
Given that a transaction looks like this:
BEGIN;
SELECT useless_field FROM useless_table WHERE useless_identifier =
'useless';
SELECT useless_field FROM useless_table WHERE useless_identifier =
'something else';
INSERT INTO useless_table (useless_field, useless_identifier) VALUES
('
Personally, I would have to disagree. I have just completed a year-long
uni project whereby we built some software that stored massive images in
InnoDB tables. We found that the performance was (at very worst)
comparable to the alternative method.
Additionally, this method is better on many levels
Really? In both cases, it's just bits on a disk. In the case where you
don't have access to a shared file repository for your client apps, you
haven't got the option of just storing paths.
Regards,
Chris
On Sat, 2003-12-13 at 23:30, Sime wrote:
> Chris Nolan wrote:
> >
Talk about a decent reply!
For web site stuff, having control over everything is pretty much a
requirement (if you want to do anything non-trivial). The way that I
serve images from the database is by parsing URLs and I've found that
the performance is very good. The fact that I can move stuff aro
Forgot something in my other reply.
With the NAS - what's to say that MySQL's retrieval and network protocol
is not more efficient than whatever is running on your NAS boxes?
Conversely, MySQL's current 16 MB per transfer limitation may very well
not allow it to act in this role at all.
Ah, the
27;ve seen for it is upgrades of custom software. It's handy
should an ALTER TABLE fail or similar. That's the sole reason that one
developer I know of deploys SQLBase to their clients.
Thanks again!
Regards,
Chris
On Sun, 2003-12-14 at 08:00, Harrison Fisk wrote:
> Hi,
>
> On
Hi!
There are plenty of funky ways to do this. :-)
The easiest and fastest way would simply need an AUTO_INCREMENT column
on your table. Then, you might be able to do something like this (with
MIGHT being the operative word):
SELECT * FROM table ORDER BY auto_inc_column DESC LIMIT 22;
I've just
Hi!
There are many ways, depending on whether you want the database to
handle it, or you want your application to handle it.
What you want is a UNIQUE index on surname_original and name_original.
Assuming the table already exists:
ALTER TABLE names ADD UNIQUE(name_original,surname_original);
Hi all, while reading through some of the MySQL docs, I noticed the
following paragraph:
|ALTER TABLE| works by making a temporary copy of the original table.
The alteration is performed on the copy, then the original table is
deleted and the new one is renamed. This is done in such a way that
Hi!
The optimizer issue is a known bug in 4.0.16 (I think). It should be
resolved soon if my memory of the traffic on the list as of late is
accurate.
Regarding the issue with threads on Redhat 9.0 - did MySQL 4.0.13-Max
come with Redhat 9.0 (or was it supplied by Redhat)? If it was, then
chances
003 at 02:58:53AM +1100, Chris Nolan wrote:
> > Hi all, while reading through some of the MySQL docs, I noticed the
> > following paragraph:
> >
> > |ALTER TABLE| works by making a temporary copy of the original table.
> > The alteration is performed on the
To elaborate on Dr Frank's thing if you're interested, here's a classic
deadlock example:
1. Transaction A obtains an exclusive lock on a set of rows which we
will call R1.
2. Transaction B obtains an exclusive lock on another set of rows which
we will call R2.
3. Transaction A requests (but ob
Hi,
As far as I know, definitely not. However, you could use an
AUTO_INCREMENT field as the independent variable
for some application-level function you use to generate the values in
the sequence.
Best regards,
Chris
Graham Little wrote:
I was wondering whether it was possible to make and AU
Huh? Not know how to backup a MySQL database? *sigh*
Every night, I do a backup of our MySQL database server that's holding
all of our mail and various other things (20GB+).
I set the isolation level to READ_REPEATABLE and use mysqldump | bzip2
to get the result. I've tested the restore and it'
Sven Köhler wrote:
I was very disappointed by Interbase/Firebird. It seemed to me like a
MS-Access: a database-engine that works on regular files
What gave you that idea? Firebird (and InterBase of course) use
a at least 1 file per database, but that's all. Can you define
"regular files"?
My i
Hi all,
I was sitting here thinking to myself (which can be quite dangerous) and
was wondering if anyone on the list actually uses dirty reads in their
apps. If so,
what advantages do you get from using this isolation level? I can't
think of any myself (damned limited brain...)
Best regards,
Jeremy Zawodny wrote:
On Mon, Dec 15, 2003 at 02:12:01PM +1100, Chris Nolan wrote:
It seems slightly ambiguous - updates are redirected and stalled. The
fact that the two statements are in different sentences threw me off
slightly.
Oh, okay. If you can suggest a more clear version
InnoDB is extremely stable!
I have a single InnoDB database that's currently holding about 20GB
(with about 95% of that in a single table).
All of this database is contained inside a single InnoDB tablespace
file. In the last 12 months, the only command
I've thrown at it by hand was "ALTER TABL
Martijn Tonies wrote:
Firebird/Interbase have all those nice things like row-level locking
(although it doesn't seem to have multiversioning like InnoDB,
PostgreSQL or Oracle), deadlock detection, prepared statements, views,
Yes it DOES have multi-versioning. Actually, I believe it was the
fi
Are they? Shoving in rows that are several meg in size didn't pose any
problems. The restore procedure looked like this:
bunzip2 dumpfile | mysql -u db_grunt -p projectdb
May I ask where the limitation you mentioned is documented? Maybe the
situations we were using it in didn't come close to th
hat's crap,
since SQL-Statements are limited in length.
On Tue, 16 Dec 2003, Chris Nolan wrote:
Are they? Shoving in rows that are several meg in size didn't pose any
problems. The restore procedure looked like this:
bunzip2 dumpfile | mysql -u db_grunt -p projectdb
May I ask
You may have to increase the size of the table cache, and you will most
probably need to do something about
ensuring that the mysqld process can open about 1 billion files at the
same time. There was a discussion along these
lines not so long ago focusing on having massive numbers of tables that
Even if your database fits entirely in memory, not having indexes in
place would not be a good idea.
In an interview Monty did regarding in-memory databases, he very
specifically made the point that where
your database is sitting will never remove the need for various types of
index.
From some
Hi all!
I was wondering if anyone on the list currently runs MySQL on Novell
NetWare. Any comments about performance relative to MySQL
on similar hardware running a different OS?
Additionally, I've been told (by someone of dubious authority) that
NetWare has (and has had for a long time) a 2 GB
table space was split up (and I'm not sure that raw partitions for
InnoDB are supported on NetWare).
Regards,
Chris
Jeremy Zawodny wrote:
On Tue, Dec 16, 2003 at 11:12:43AM +1100, Chris Nolan wrote:
Hi all!
I was wondering if anyone on the list currently runs MySQL on Novell
NetWare.
Hi,
Have you considered just normalising the schema a bit more? It looks
like you're attempting to duplicate data within the
table (which you don't strictly need) and duplicating Fname and Lname
between Blah and foo. Why not just have an
AUTO_INCREMENT column (or some other unique row identifier
Hi,
Regarding backups, mysqlhotcopy locks all MyISAM tables named for backup
before copying the files. As a result, they are in a consistant state.
InnoDB and BDB tables need to be handled differently though, so you're
looking at 1 of 4 methods:
1. Shut the database down and copy the table spa
Curtis Maurand wrote:
Matthew Stanfield said:
Hi,
Usually, i'll use "enum('0','1')" in place of a boolean type.
Curtis
[snip]
For JDBC stuff, I've found that if you really want to call this a
shortcoming, then that's about as far as you can take it
- the MySQL JDBC driver makes the B
mos wrote:
At 04:22 AM 12/15/2003, you wrote:
To elaborate on Dr Frank's thing if you're interested, here's a
classic deadlock example:
1. Transaction A obtains an exclusive lock on a set of rows which we
will call R1.
2. Transaction B obtains an exclusive lock on another set of rows
which we
There are a few things that could be happening:
1. Solaris 8 uses many-to-many threads in "Solaris threading mode" and
1:1 in
POSIX threading mode. I'm not sure which functionality the MySQL
binaries exercise,
but 1:1 seems to be less intensive. Regardless, Solaris thread creation
tends to be s
Dr. Frank Ullrich wrote:
Hi!
--cut
To be honest, the vast majority of database installations experience
problems in performance caused by poor query and schema design,
bad application logic or grossly underspecified hardware. A change in
the number of files used to store the data is extremely
Hi!
How heavy is your usage of TEMPORARY TABLES? I don't use them much
myself, but
I'm sure that the others on the list will have something to say in that
regard.
To get a better look at MySQL's usage of memory, you could try looking
at the output of
SHOW STATUS .
Regards,
Chris
Markus Fisc
Jeremy Zawodny wrote:
On Tue, Dec 16, 2003 at 12:01:55PM +0700, David Garamond wrote:
Sven K?hler wrote:
I was very disappointed by Interbase/Firebird. It seemed to me like a
MS-Access: a database-engine that works on regular files
Firebird seems simple, but it doesn't mean it's in
Matthew Stanfield wrote:
Usually, i'll use "enum('0','1')" in place of a boolean type.
Curtis
For JDBC stuff, I've found that if you really want to call this a
shortcoming, then that's about as far as you can take it
- the MySQL JDBC driver makes the BIT field act just like a
single-bit field
mos wrote:
At 01:14 AM 12/16/2003, you wrote:
mos wrote:
At 04:22 AM 12/15/2003, you wrote:
To elaborate on Dr Frank's thing if you're interested, here's a
classic deadlock example:
1. Transaction A obtains an exclusive lock on a set of rows which
we will call R1.
2. Transaction B obtains an
David Griffiths wrote:
To be honest, the vast majority of database installations experience
problems in performance caused by poor query and schema design,
bad application logic or grossly underspecified hardware. A change in
the number of files used to store the data is extremely unlikely
to reso
Chris Allen wrote:
On Tue, Dec 16, 2003 at 10:13:39PM +1100, Chris Nolan wrote:
There are a few things that could be happening:
Are you using persistant connections from your web servers,
Yes - with Apache::DBI under mod_perl
Well, that's one option out of the way,,,
2. Whi
32-bit filesystem limits?
Looking at any modern Linux FS, your file size limits are not hindered
by 32-bit anything or even
the FS itself. On kernel 2.4, internal kernel structures limit the
maximum size of block devices to around 1 TB.
As a result, you can "only" have files of about that size (
InnoDB does "clustering in the Sybase style" but MyISAM is lean, mean
and quick.
If you really wanted to do this though, you could probably do the
following:
CREATE TABLE tab SELECT * FROM old_tab ORDER BY zip_code ASC;
And then add the indexes!
One thing you should know though - MyISAM's index
Hi all,
Upon reading the funky manual, I have discovered the following things:
1. TIMESTAMP fields can be set so that their default value is NOW().
2. DATE and TIMESTAMP fields are related.
Given the two above facts, is there a way to set DATE columns so the
default value is NOW()? My playing ar
;t want it update
> when you UPDATE the row, you can just set it to its current value, if
> you weren't aware of that.
>
>
> Matt
>
>
> - Original Message -
> From: "Chris Nolan"
> Sent: Saturday, January 03, 2004 10:34 AM
> Subject: Defau
On Thu, 2004-01-08 at 21:10, [EMAIL PROTECTED] wrote:
> Andy Bakun wrote:
>
> >On Wed, 2004-01-07 at 10:38, [EMAIL PROTECTED] wrote:
> >
> >
> >>Hi all,
> >>does anyone know if it's possible to compile MySQL under Linux so that
> >>mysqld doesn't rely upon LinuxThreads, but makes direct call to
On Fri, 2004-01-09 at 20:58, Andy Bakun wrote:
> On Thu, 2004-01-08 at 05:24, Chris Nolan wrote:
>
> > 3. Wait for a while. Linux 2.6 includes (as does the RedHat 9.0 and
> > ES/WS/AS 3.0 kernels) NPTL - the Native POSIX Threads for Linux
> > implementation which is superi
This is a fuzzy issue.
There are questions regarding redistribution. Any distribution requires
that you either comply with the terms of the GPL or that you get a
licence.
Additionally, MySQL AB have recently changed the licence terms of their
libraries - now absolutely everything that the fine an
"
MIME-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 8bit
Hmm
Will you have an object hierarchy of any type to speak of? If so, you
should be able to factor quite a few things out, but then you're going
to have the problem that objects further down the tree will take longer
Depending on your language, you might be able to "fudge" this up to work
for yourself. In C/C++ I have a function that automagically decides
whether I am beginning a new transaction when I am calling it or whether
I want Oracle-style nested transaction functionality. My rollback
function is aware o
"
MIME-Version: 1.0
Content-Type: text/plain
Content-Transfer-Encoding: 8bit
You'll find that the problem is probably related to the new password
format that the 4.1.x stream of MySQL uses. You can start the server
with some switch (listed in the manual) that allows use of old
passwords.
Regards,
Hi Alex!
On Sat, 2004-01-17 at 05:50, Zeltser, Alex wrote:
> Hi,
>
> I wanted to take advantage of the InnoDB 'gap' locking to lock 'non-existence' of a
> row, the way the
> manual recommends. I tried to do this by using 'select ... for update', using the
> 'mysql' client
> from two separate s
Hmm...have you looked at Rekall?
www.total-rekall.co.uk
Also, you might want to check out OpenOffice.org's database interface
features
Regards,
Chris
On Sat, 2004-01-17 at 11:12, Bryan Koschmann - GKT wrote:
> I wanted to thank everyone for their responses and information regarding
> this. I a
Hi!
You're looking for the function my_free(). Enjoy!
Regards,
Chris
John McCaskey wrote:
I have the following code:
//try the mysql connection
mysql_init(&mysql_connection);
if(!mysql_real_connect(&mysql_connection, db_host, db_user, db_pass,
db_db, 0, NULL, 0)) {
Hi all,
I've been away from the list for a while and am wondering if the
following question has been answered:
Given that the extremely funky InnoDB is going to gain a new file format
in the future, would it be a fair guess to say that any additions to
InnoDB requiring file format changes would b
Hi all,
Does anyone on the list have experience storing their tables and
associated bits and pieces on raw partitions? The reason I ask is two fold:
1. Did you notice much in the way of performance difference (which OS
you're using is probably very relevant here).
2. Have you had to deal with h
MySQL can help you out here, but there are many questions you need to
answer first.
1. What exactly is in the DOC column? Binary data of some disgusting,
inferior proprietary word
processor document format, plain text, XML, LaTeX files?
2. If you answered "But I like Clippy!" to the above, go to
Hmmfor practical purposes:
1. MySQL is going to cost you a lot less, no matter which way you do things.
2. MySQL is going to perform better for the vast majority of workloads.
The only place where MS SQL Server *might* have an advantage is in
situations where it's additional language feature
Hi!
If memory serves me correctly, MySQL AB have stopped issuing SCO
OpenServer and SCO UnixWare binaries publicly and will only be providing
them to customers who take out commercial licences. This is probably due
to the fact that OpenServer and UnixWare cost a lot compared ot other
OSes, don
Hi all!
I remember reading a number of articles on Jeremy Zawodny's website
regarding FreeBSD and Linux performance as it relates to MySQL. At the
time of publication, it seemed that Linux's native thread library had
quite an advantage over FreeBSD's (but FreeBSD included LinuxThreads
support
Hello Martijn!
It seems that whenever we both comment in a thread, you enlighten me
greatly!
Martijn Tonies wrote:
Hi Chris,
I understand that you like MySQL but ...
Hmmfor practical purposes:
1. MySQL is going to cost you a lot less, no matter which way you do
things.
This is a
Martijn Tonies wrote:
Hi Chris,
It seems that whenever we both comment in a thread, you enlighten me
greatly!
;-) ... I'm learning more about MySQL with every post. Ok, maybe
not every post, but still ... *g*
I tend to be a critic sometimes, but I'm a really nice guy. Believe me on
this o
Anup Mahansaria
Chris Nolan
<[EMAIL PROTECTED]
>
Hi!
For formatted text, you may be able to get away with using FULLTEXT
searches on MyISAM tables, depending on the definition of "formatted".
Regards,
Chris
On Wed, 2004-02-11 at 18:55, Veysel Harun Sahin wrote:
> Hello list,
>
> What is the best way to store and search formatted text?
>
> T
On Wed, 2004-02-11 at 22:29, Jochem van Dieten wrote:
> Chris Nolan wrote:
> > Martijn Tonies wrote:
>
> > Additionally, it is an accepted fact that MySQL is faster than the
> > mighty, mighty PostgreSQL.
>
> No, it is not. It is an accepted fact that MySQL is
dingly).
Regards,
Chris
Ed Leafe wrote:
On Feb 10, 2004, at 9:12 AM, Chris Nolan wrote:
12. MySQL AB weren't responsible for afflicting the world with the
Jet database engine (Access) or Visual FoxPro, thus they are more
trustworthy than MS! :-)
Microsoft *bought* FoxPro; they didn&
Hi all,
I'm looking at developing an (open source) server-style application with
the embedded MySQL library to be employed as the primary data store.
Has anyone attempted to use InnoDB Hot Backup for such a beast? I ask as
I do not know whether communication between ibbackup and the database
e
Ed Leafe wrote:
On Feb 11, 2004, at 7:31 PM, Chris Nolan wrote:
Yes, we all know that Microsoft *bought* FoxPro's underlaying
technology, that is *FoxBASE*! Everything ever called FoxPro has been
a Microsoft product.
Sorry, you're off by a few years. FoxPro had been out f
Michael McTernan wrote:
Hi,
SELECT COUNT(*) for InnoDB tables is a know problem... The table
handler (for InnoDB) has to do a table scan to count all rows... This
particular case is optimized with MyISAM ...
Sure. But why is the tablescan ~100 times faster for the table without the
BLO
n the challenge or run
screaming in the other direction and have a somewhat easier year.
Regards,
Chris
Heikki Tuuri wrote:
Chris,
- Original Message -
From: "Chris Nolan" <[EMAIL PROTECTED]>
Newsgroups: mailing.database.myodbc
Sent: Thursday, February 12, 2004 2:45 A
First and foremost, your English is not even remotely "bad"! You should
hear half of my native-English speaking friends!
Can you give us some more information, such as the server configuration,
OS, filesystem, MySQL version, table types in use, table size, size of
the data gone missing, backup
Hi!
You might want to keep your eye on MySQL Cluster - to be demonstrated at
the 2004 MySQL conference, where you might get a chance to swim with the
dolphins!
Regards,
Chris
Are Pedersen wrote:
Hello
I am looking for software/hardware solutions to provide loadbalancing
and redundancy for
substrings in the TEXT column for rows between two dates with
certain values for the ENUM columns. There are indexes on the ENUM,
TIMESTAMP and TEXT columns (prefix indexes of course). Unless I specify
FORCE INDEX(date), MySQL does a full table scan, resulting in queries
that take 2 minutes t
Hi David,
David Griffiths wrote:
From reading the docs, a binary log is an efficient representation of all
data-modifying SQL that is run on the master database. I was unable to
figure out what happens if a slave is interrupted while in the middle of
processing a binary log.
When a binary log is
files as they are reading them
in so as not to violate any constraints?
David
- Original Message -
From: "Chris Nolan" <[EMAIL PROTECTED]>
To: "David Griffiths" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Friday, February 13, 2004 5:46 PM
Subject:
Hmm
I'd check for hard disc problems (turn on SMART monitoring and look
through your kernel logs), RAM problems (look at memtest) and rootkits.
Regards,
Chris
On Sun, 2004-02-15 at 15:45, Juan E Suris wrote:
> I just installed 4.0.16 linux (x86, libc6) rpms on a fresh RH7.3 installation. I
Hi all,
I'm currently designing an open-source messaging server that will use
MySQL as the data store (in embedded form).
High performance is one of the goals of this project, so I have been
examining possible I/O models and seem to have settled on a model where
each thread services many requests
Ahh, the infamous JDBC benchmark. :-)
There's been much commentary on this in the past. There are some things
to consider:
1. The native APIs for each DB will be faster.
2. The DB, App Server and Web Server were all on one box.
3. I'm not sure if the MySQL JDBC driver supported the NamedPipeFacto
On Sat, 2004-02-14 at 08:13, Heikki Tuuri wrote:
> Chris,
>
> - Original Message -
> From: "Chris Nolan" <[EMAIL PROTECTED]>
> To: "Heikki Tuuri" <[EMAIL PROTECTED]>
> Cc: <[EMAIL PROTECTED]>
> Sent: Thursday, February
Hi!
Next-key locking essentially doesn't work on rows - it works on indexes.
It ensures that "phantom reads" can't happen.
InnoDB does indeed do row-locking. In fact, it has one of the most
efficient representations of locks of any relational database.
Regards,
Chris
On Mon, 2004-02-16 at 23:4
Hi!
MySQL's optimizer has a slight problem. OR queries cause it to get very
confused.
Try the following to get the best performance:
Rewrite SELECT FROM table WHERE (condition1) OR (condition2);
As:
(SELECT FROM table WHERE condition1) UNION (SELECT FROM table WHERE
condition2);
Hope this hel
Hello again!
Andreas Pardeike wrote:
On 2004-02-16, at 14.13, Chris Nolan wrote:
MySQL's optimizer has a slight problem. OR queries cause it to get very
confused.
Try the following to get the best performance:
Rewrite SELECT FROM table WHERE (condition1) OR (condition2);
As:
(SELECT
eployable and an original open source work.
Regards,
Chris
Curtis Maurand wrote:
checkout http://www.dbmail.org
On Sun, 15 Feb 2004, Chris Nolan wrote:
Hi all,
I'm currently designing an open-source messaging server that will use
MySQL as the data store (in embedded form).
High perf
James
-Original Message-
From: Chris Nolan [mailto:[EMAIL PROTECTED]
Sent: Monday, February 16, 2004 4:52 AM
To: James Kelty
Cc: [EMAIL PROTECTED]
Subject: Re: Locking
Hi!
Next-key locking essentially doesn't work on rows - it works on indexes.
It ensures that "phantom reads" can&#
You might be able to cheat and replicate the required database to the
local machine.
Regards,
Chris
Terence wrote:
you'll need to create temporary tables in one of the servers based on the
results of the other and then join.
- Original Message -
From: <[EMAIL PROTECTED]>
To: <[EMAIL P
servers to
one server? many masters and one slave, the slave might have replicas of
each master database
(cause i need to query on many diferent mysql servers)
thanks...
FBR
Chris Nolan <[EMAIL PROTECTED]>
17/02/2004 12:52 a.m.
To
Terence <[EMAIL PROTECTED]>
cc
[EMAIL PROTECTED]
Subje
Hi!
You'll want to look through the MySQL C API docs. There are specific
calls for allocating certain structures that MySQL client software
should use (and routines for deallocating it as well).
Regards,
Chris
Sp.Raja wrote:
Hi List,
I want to use some other memory allocator routine other t
Michael McTernan wrote:
Hi there,
Given this model, each thread is obviously going to want to have
multiple transactions outstanding. Is this something that might be added
to MySQL in future or am I totally overestimating the expense of using
one thread per connection?
I'm guessing you me
ssage
From: Chris Nolan <[EMAIL PROTECTED]>
To: "Sp.Raja" <[EMAIL PROTECTED]>
Cc: [EMAIL PROTECTED]
Date: Tue, Feb-17-2004 6:29 PM
Subject: Re: MySQL Memory
Hi!
You'll want to look through the MySQL C API docs. There are specific
calls for allocating certain struc
COUNT(*) is a special case for MyISAM. However, you'll find that
anything that has a WHERE clause that takes advantage of an index is
pretty quick for both MyISAM and InnoDB tables.
For instance:
SELECT COUNT(*) FROM table;
Is slow as all buggery on InnoDB, but:
SELECT COUNT(id) FROM table WH
Craig Robinson wrote:
Hi,
I intend to use MySQL for a database application, and I am not sure
whether to use MyISAM or InnoDB tables. There will be a very large
amount of (mainly numerical) data. The relations between the tables will
be reasonably simple. Which table type would you recommend?
The
Simon Green wrote:
Is it just not the case that InnoDB table have to do more as they have more
functionality and so take more time?
Not exactly.
InnoDB does indeed support transactions, uses the ultimate in
concurrency control (multiversioning) and provides foreign key constraints.
That said
Hi!
Eric B. wrote:
Hi,
I'm trying to implement the proper indexing for my DB, and am having some
challenges. I was hoping someone could lend me a hand.
If I have 5 fields in the DB that I am indexing (field1, field2, field3,
field4, and field5), I know I can create an index:
Key Index1( field1,
Andy Bakun wrote:
On Thu, 2004-02-19 at 08:24, Alex Greg wrote:
I have a select query which is selecting all records in a table (which has
around 8,000,000 rows in). "time" is a field of type "time". Should I be using
<= and >= or BETWEEN to find records in a certain range? Which does MySQL
op
Steve Edberg wrote:
It's my understanding that doing a simple delete
delete from table_name
actually DOES drop and recreate the table (and thus its indexes). On
the other hand, if you are continually adding & deleting records, you
might well need to do a periodic 'analyze table_name' o
Sasha Pachev wrote:
C versus object-oriented lanuguages like C++/Java is a topic I have
discussed a lot with programmers. I believe that traditional procedural
approaches and languages, like C, are the best for 'systems
programming', by
which I mean implementing anything with complex data struct
Sasha Pachev wrote:
Jochem van Dieten wrote:
Sasha Pachev wrote:
Heikki Tuuri wrote:
C versus object-oriented lanuguages like C++/Java is a topic I have
discussed a lot with programmers. I believe that traditional
procedural
approaches and languages, like C, are the best for 'systems
programm
Hmm...if there's lots of thrashing, it might be to do with
fragmentation. Have you tried running OPTIMIZE TABLE on the table in
question?
Does anyone on the list have anything to say about putting the MYD and
MYI files on seperate disks or using RAID MyISAM tables??
Regards,
Chris
Eric B. wr
Can you send us the CREATE TABLE statement for this troublesome table?
Regards,
Chris
Eric B. wrote:
Sorry - forgot to mention it. I've already tried both an OPTIMIZE TABLE and
ANALYZE TABLE to try to improve performance, but with no result.
Any other ideas?
Thanks,
Eric
"C
On Sat, 2004-02-21 at 20:39, Franz, Fa. PostDirekt MA wrote:
> Hi,
>
> this discussion is useless, object or procedure is not realy the question.
> You need to know how to build a good programm, if you cannot create a good programm,
> no matter what language.
Originally, I was asking Heikki's opin
I don't like your chances of successfully doing this. You could try
pulling the changesets out of the bitkeeper repository by hand but the
fact that 4.1.X was branched from an earlier version of the 4.0.x series
(I think), it might be a bit of a strugle.
Anyways, good luck! :-)
Regards,
Chris
O
If there's a bug in the optimiser, you'll find it's in the "higher
levels" of the codebase. InnoDB doesn't have any SQL optimisation code
in it, so any bugs in this area aren't in InnoDB.
Regards,
Chris
Dyego Souza Dantas Leal wrote:
The script of database are uploaded to:
support.mysql.com/
Hmm
The question is, does MySQL's optimiser do enough planing to result in a
tree of any non-trivial interest?
I love MySQL as much as the next geek with a significant other that
loves dolphins, but I'm not sure that MySQL 4.0 would provide a lot of
data for funky tree-drawing (MS SQL tool
1 - 100 of 158 matches
Mail list logo
