Hi Geoffrey,
I beg to disagree. A simple redesign should do the trick. The problem IMHO
is that the query in question has to go through the whole database each
time. Busting the table up into smaller tables ordered by date keys and
categories or whatever with keys to the articles should do the trick with
MySQL and modest hardware.
- yba
On Wed, 21 Mar 2007, Geoffrey S. Mendelson wrote:
Date: Wed, 21 Mar 2007 16:27:56 +0200
From: Geoffrey S. Mendelson <[EMAIL PROTECTED]>
To: Elazar Leibovich <[EMAIL PROTECTED]>
Cc: linux-il@linux.org.il
Subject: Re: [Job] MySQL consultation
On Wed, Mar 21, 2007 at 03:42:21PM +0200, Elazar Leibovich wrote:
It seems that my question boils down to this:
I'm having a 3Gb table with three TEXT columns and an ID.
Counting it takes between 1 to 2 minutes, is it normal for such a size
of a table?
I think it's pretty obvious that you are pushing the practical
limits of MySQL and PC type hardware. Reading a 3gb table in two minutes
if pretty fast, that's 25 megabytes a second, assuming you are really
reading the table and not an index, which is more likely.
If you really want to have a database that large and do anything
except minor record insertion, updates and retrieval, you are IMHO
a canditate for Oracle or DB2 with the appropriate hardware and operating
system.
You should plan for the maximum size of your database, double it unless
you are 100% sure your estimates are perfect and then choose the
technology to match.
There was someone on this list a week or two ago looking for a job
who had the skills you need now and in the future. You probably
should hire him now, before you spend a lot of time and money
and have to retool because you made a poor choice.
Geoff.
--
EE 77 7F 30 4A 64 2E C5 83 5F E7 49 A6 82 29 BA ~. .~ Tk Open Systems
=}------------------------------------------------ooO--U--Ooo------------{=
- [EMAIL PROTECTED] - tel: +972.2.679.5364, http://www.tkos.co.il -
=================================================================
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word "unsubscribe" in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
