From owner-freebsd-hackers Tue Jan 26 12:43:49 1999 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA27118 for freebsd-hackers-outgoing; Tue, 26 Jan 1999 12:43:49 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from feral-gw.feral.com (feral.com [192.67.166.1]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA27110 for ; Tue, 26 Jan 1999 12:43:45 -0800 (PST) (envelope-from mjacob@feral.com) Received: from localhost (mjacob@localhost) by feral-gw.feral.com (8.8.7/8.8.7) with ESMTP id MAA20692; Tue, 26 Jan 1999 12:43:02 -0800 Date: Tue, 26 Jan 1999 12:43:02 -0800 (PST) From: Matthew Jacob X-Sender: mjacob@feral-gw Reply-To: mjacob@feral.com To: David Wolfskill cc: freebsd-hackers@FreeBSD.ORG Subject: Re: Error in vm_fault change In-Reply-To: <199901262036.MAA08253@pau-amma.whistle.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG It would be a good thing to add knobs back in to try and give some control over interactive vs. single task throughput performance. Lacking knobs, configuration options would be good. > >From: Robert Watson > > >I.e., today it's a realtime machine, tomorrow it's a single-user > >workstation, the next day a heavily loaded timesharing machine with CPU > >time partitioning. > > I can see a value to this ability.... > > Back when I was an MVS (IBM mainframe) systems programmer, it wasn't > uncommon for a machine to be set up to have differing SRM (? "System > Resource Manager" seems to strike a resonant chord in my memory) > "objective curves" depending on the expected workload, which would often > vary depending on time-of-day: for example, during the day, the system > would be weighted to favor interactive processes, while at night, it > would switch to more of a batch mode of operation (for back-office > file-crunching). > > Although the range of work handled reasonably adequately by modern UNIX > systems is pretty awesome, the ability to favor one type of workload > over another can give a sysadmin another "knob" to twist -- and it may > merely be used more on a per-site basis than on a time-of-day basis, but > it would be rather silly to assume that everyone's workload is the same. > > The boxes I'm responsible for here, for example, aren't under the kind > of load (qualitatively or quantitatively) that the shellN boxes at BEST > (Hi, Matt!) are, for example. (I consider this A Good Thing.) And the > workload of one box may well differ from that of another; this is > intentional. > > david > -- > David Wolfskill UNIX System Administrator > dhw@whistle.com voice: (650) 577-7158 pager: (650) 371-4621 > > To Unsubscribe: send mail to majordomo@FreeBSD.org > with "unsubscribe freebsd-hackers" in the body of the message > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message