Skip site navigation (1)Skip section navigation (2)
Date:      Wed, 15 May 2002 21:45:21 -0400 (CLT)
From:      Roberto de Iriarte <roberto@spock.cl>
To:        Diego Montalvo <dmontal2@yahoo.com>
Cc:        alpha@FreeBSD.ORG
Subject:   Re: Alpha CPU Performance vs i386
Message-ID:  <Pine.BSF.4.21.0205152106190.28379-100000@aromo.spock.cl>
In-Reply-To: <20020515225520.79881.qmail@web13408.mail.yahoo.com>

next in thread | previous in thread | raw e-mail | index | archive | help


On Wed, 15 May 2002, Diego Montalvo wrote:

> 
>  Howdy,
> If I am running a database intensive program such as mysql, on a 500MHZ Alpha, and a 1GHZ PIII, which would preform faster? 
> Diego 
>  

It depends, of course. Quickly browsing the Spec CINT2000 Benckmark
suite, i'd say, for most mysql usage, 197.parser and 255.vortex, might
give you an idea of relative performance. Note that the system
configurations used for this benchmarks are highly optimized, but, just
for the comparison, let's think that the vendors's optimizations even out
each other. (I will use CINT200-BASE numbers)

An Intel PIII 1.0 Ghz

Intel VC820 (Rambus Chipset) 133 Mhz FSB

197.parser 340
255.vortex 706

Alphas at 500 Mhz

An oldie (AlphaStation 500/500) 21164A 8 Mb cache

197.parser 130
255.vortex 207

Also an oldie (DEC Personal Workstation 500 AU) 21164A 2 Mb cache

197.parser 124
255.vortex 203

Not that old (Dec-98) (AlphaServer DS20 Model 6/500) 21264 4 Mb cache

197.parser 206
255.vortex 400

So, i'd say an Intel PIII would be faster than an Alpha if your database
application performance would be CPU bound. That is usually not the case
(IBM Mainframes would have been given up a long time ago if it were!), so
it would depend on the particular configuration of your systems. A big 
reason why old Alphas, Sun's, RS/6000's, etc have a long useful life
lies in the good design of the IO subsystem and peripherals (Not multia
of course :)) For that, you must think, in PC terms, about higher end
server class machines that are not that cheap. 

Regards
Roberto

BTW, Reading the detailed results of the various SPEC benchmarks is
much more usefull than reading the baseline number.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-alpha" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.4.21.0205152106190.28379-100000>