*BSD News Article 91926


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!news.mira.net.au!pumpkin.pangea.ca!www.nntp.primenet.com!nntp.primenet.com!news.mathworks.com!enews.sgi.com!news.corp.sgi.com!fido.asd.sgi.com!blit.engr.sgi.com!nate
From: nate@blit.engr.sgi.com (Nate Tuck)
Newsgroups: comp.unix.bsd.freebsd.misc,comp.sys.sgi.misc
Subject: Re: no such thing as a "general user community"
Date: 26 Mar 1997 01:51:57 GMT
Organization: Silicon Graphics, Inc., Mountain View, CA
Lines: 30
Message-ID: <5h9vft$8eo@fido.asd.sgi.com>
References: <331BB7DD.28EC@net5.net> <5h5jgr$et4$1@news.clinet.fi> <5h91l2$gua@innocence.interface-business.de> <5h9rr0$2sj@flea.best.net>
NNTP-Posting-Host: blit.engr.sgi.com
Xref: euryale.cc.adfa.oz.au comp.unix.bsd.freebsd.misc:37722 comp.sys.sgi.misc:29399

In article <5h9rr0$2sj@flea.best.net>,
Matt Dillon <dillon@flea.best.net> wrote:
>    At the moment, we are getting an average of 20 to 30 days uptime on 
>    our L's, running 64 bit 6.2 with a generally heavy multi-user load, 
>    and 60 to 100 days or so of uptime on the Indy's and S's which 
>    are dedicated to particular functions (such as DNS, irc, and sendmail). 
>    The challenge S's with medium to heavy multi-user loads tend to stay up
>    around 40 days.   Sometimes the kernels get confused and have to
...
>    The FreeBSD machines tend to stay up for months (100+ days with a medium
>    to heavy user load, or a heavy dedicated load) and basically not die 

I think we can all argue about stability until we are blue in the
face, the fact is there are bugs in every OS.  Sometimes you find them
(and get pissed off) and sometimes somebody else finds them for you.
But, you've started to talk about something I actually wonder
about...

How do you define these load metrics?  What is a heavy load on each of
the machines?  Is it measured by xload, response time on some app, or
what?  How many users can you stick on each machine before the load
becomes heavy?

Religious differences and rose-colored glasses aside, what is the
difference in throughput between the two platforms in the specific
case of BEST?  Where does SGI need to do some code tuning?

I'm interested.

nate