From owner-freebsd-chat  Fri Aug 29 08:46:14 1997
Return-Path: <owner-freebsd-chat>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id IAA10629
          for chat-outgoing; Fri, 29 Aug 1997 08:46:14 -0700 (PDT)
Received: from implode.root.com (implode.root.com [198.145.90.17])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id IAA10621
          for <chat@FreeBSD.ORG>; Fri, 29 Aug 1997 08:46:09 -0700 (PDT)
Received: from implode.root.com (localhost [127.0.0.1])
	by implode.root.com (8.8.5/8.8.5) with ESMTP id IAA18611;
	Fri, 29 Aug 1997 08:48:29 -0700 (PDT)
Message-Id: <199708291548.IAA18611@implode.root.com>
To: Nate Williams <nate@mt.sri.com>
cc: chat@FreeBSD.ORG
Subject: Re: Dual Log Stats - 1997/08/29 
In-reply-to: Your message of "Fri, 29 Aug 1997 09:14:45 MDT."
             <199708291514.JAA08864@rocky.mt.sri.com> 
From: David Greenman <dg@root.com>
Reply-To: dg@root.com
Date: Fri, 29 Aug 1997 08:48:29 -0700
Sender: owner-freebsd-chat@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

>> limit. It would be nice if we had enough reserve capacity to handle special
>> days like yesterday. I estimate that we could have handled everyone (not
>> turned anyone away) if the limit had instead been around 3000. The main
>> problem is lack of main memory, but we're pretty much out of CPU cycles at
>> 2000 users as well.
>
>*chanting* SMP, SMP, SMP...
>
><grin>

   With the work that Steve is doing right now in leveraging the spl*
functions to allow CPUs to process interrupts simultaneously while another
CPU is executing system call code, going to SMP on wcarchive is a definate
possibility. I'd say this would be at least four months out. however.
Basically, the decision to go SMP on wcarchive is simple: It has to help
more than it hurts.

-DG

David Greenman
Core-team/Principal Architect, The FreeBSD Project