[prev in list] [next in list] [prev in thread] [next in thread] 

List:       mysql-internals
Subject:    Re: Missed optimization in 3.23.33
From:       "Paul Cadach" <paul () odt ! east ! telecom ! kz>
Date:       2001-08-21 13:13:07
[Download RAW message or body]

Hi,

----- Original Message -----
From: "Michael Widenius" <monty@mysql.com>
To: "Paul Cadach" <paul@odt.east.telecom.kz>
Cc: <internals@lists.mysql.com>
Sent: Tuesday, August 21, 2001 7:55 PM
Subject: Missed optimization in 3.23.33


> Paul> May be this optimization requres to bring weighting for functions,
to
> Paul> execute simplest functions first, then (if condition is true) more
complex,
> Paul> and last (if condition still true) - most complex functions? This
weight may
> Paul> be a value of time which each function executed 10000 (for example)
times.
>
> The problem is that it's almost impossible for the optimizer to say
> which function will be faster, as this will depend on the arguments
> and the data.  When executing expression MySQL does most things from
> left to right, so it's relatively easy for a human to get MySQL to do
> the right thing.

The problem was solved by using of weightings for comparsion functions at
application level when SQL expression builds.


WBR,
Paul.


---------------------------------------------------------------------
Before posting, please check:
   http://www.mysql.com/manual.php   (the manual)
   http://lists.mysql.com/           (the list archive)

To request this thread, e-mail internals-thread1517@lists.mysql.com
To unsubscribe, e-mail <internals-unsubscribe@lists.mysql.com>

[prev in list] [next in list] [prev in thread] [next in thread]