On Tue, Jan 16, 2001 at 01:43:26AM -0800, Derek Sivers wrote:
> Has anyone done testing to show the difference between:
>
> #1 - having MySQL on the same server as Apache/PHP
>
> vs.
>
> #2 - having MySQL on its own dedicated computer, TCP/IP connected to the
> main Apache box
>
> NOTE: We use lots of little MySQL lookups in our PHP scripts. Nothing
> that intensive. But our site is getting pretty MySQL-intensive and often
> MySQL takes up 30% of the CPU power.
The first option simply won't scale beyond a certain point (assuming
limited hardware resources). You *could* go get an E-10000 from Sun
and spend a quarter of a million dollars, but you can probably get by
with splitting things up a bit.
Assuming your machines are "close" to each other (network-wise), you
probably won't see much of a difference. All of my hard-core database
stuff is done over TCP/IP to a dedicated server and the performance is
quite good (they're on 100Mbit switched ethernet).
But your mileage may vary. You'd need to test that.
We're in the midst of some testing involving a replicated database
running on the same machine as some hard-core number crunching where
the latency of fetching the data is fairly important. So far things
look good, but we really haven't stressed it yet. I'm anxious to see
how our testing turns out.
Jeremy
--
Jeremy D. Zawodny, <[EMAIL PROTECTED]>
Technical Yahoo - Yahoo Finance
Desk: (408) 328-7878 Fax: (408) 530-5454
Cell: (408) 439-9951
---------------------------------------------------------------------
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 <[EMAIL PROTECTED]>
To unsubscribe, e-mail <[EMAIL PROTECTED]>
Trouble unsubscribing? Try: http://lists.mysql.com/php/unsubscribe.php
