From owner-freebsd-hackers Thu Mar 7 10:55:53 2002 Delivered-To: freebsd-hackers@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id 17C5337B400 for ; Thu, 7 Mar 2002 10:55:45 -0800 (PST) Received: (from dillon@localhost) by apollo.backplane.com (8.11.6/8.9.1) id g27Ith468535; Thu, 7 Mar 2002 10:55:43 -0800 (PST) (envelope-from dillon) Date: Thu, 7 Mar 2002 10:55:43 -0800 (PST) From: Matthew Dillon Message-Id: <200203071855.g27Ith468535@apollo.backplane.com> To: Tom Cc: Dimitar Peikov , cjp , Subject: Re: Swapping performance References: <20020307095452.D18855-100000@frond.minions.com> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG :... :> > This is a comparison of how fast Linux can do something :> > STUPID versus how fast a real OS can do something intelligently. Your :> > test is giving you misleading, and dangerous numbers. Do not go waving :> > them around until you have actually looked at mallocs behavior on the :> > different systems. :> :> In fact if I have to compute something really important for me (STUPID as :> you said) I would choose the fastest OS. : :But when you lose that data, do you not get burnt by that same situation? :I have written a 1GB file to a linux box, and then within 5 seconds of it :finishing, yanked the power cord. When I booted it back up, the file was :*JUST NOT THERE*, I tried it a few other times, and there were fragments :that showed up. Under FreeBSD I tried the same test, The file was there, :and it finished faster than Linux did. Why is this? Bad procedure to gain :file system speed (from what I can tell). : :Which would you rather have? Fast Calculations, or the results of your :data. Obviously its your choice :) :... :--- :Tom bifrost@minions.com Well, even under FreeBSD yanking the plug can lead to the just-written file being lost. It depends what filesystem options you use. And yanking the plug would also be dangerous if the hard disk were doing a write operation at the time of the yank no matter the OS. Then you'd lose some sectors (or even an entire track, or multiple tracks depending the drive manufacturer). The only reason this particular test worked for you is probably because you were writing the file out sequentially and FreeBSD does clustered write-behind and file-cache-invalidate-behind when a file is being written out sequentially. But this is just one particular feature under the specified conditions so you can't really use it to justify a general comparison. -Matt Matthew Dillon To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message