Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!uunet!kddlab!titcca!cc.titech.ac.jp!necom830!mohta
From: mohta@necom830.cc.titech.ac.jp (Masataka Ohta)
Newsgroups: comp.arch
Subject: Re: Sun bogosities, including MMU thrashing
Message-ID: <7191@titcce.cc.titech.ac.jp>
Date: 7 Feb 91 02:40:39 GMT
References: <5257@auspex.auspex.com> <3956@skye.ed.ac.uk> <PCG.91Jan23183334@odin.cs.aber.ac.uk> <399@appserv.Eng.Sun.COM> <PCG.91Feb4191423@odin.cs.aber.ac.uk> <417@appserv.Eng.Sun.COM>
Sender: news@cc.titech.ac.jp
Organization: Tokyo Institute of Technology
Lines: 30

In article <417@appserv.Eng.Sun.COM> lm@slovax.Eng.Sun.COM (Larry McVoy) writes:

>Subject: Re: FLAME ON: Re: Sun bogosities, including MMU thrashing

>Ready flame throwers, take aim, .... fire!

Would you stop posting such things unless you want to be the living
evidence of Sun bogosity?

>Since I kept the read ahead alg, I'm now
>doing a 111KB of useless read by your definition.  (You seem to be assuming
>that 1KB is some nifty magic size.)

Everyone will admit it work very well 80% of time.

>>Let me guess: a large, large percentage of your 4KB-8KB buffers or pages
>>that are used for IO contain files or directories whose median size is
>>well under twice that of a page, and a fairly large percentage contain
>>files that are smaller than half a page.

>A little worse, but notice that lack of a knee at 1KB.

No one claimed there is a knee at 1KB.

						Masataka Ohta

PS

I prefere a good (or even bad) armchair architect much better than a bad
architect developping some real product.