From owner-freebsd-stable Thu Feb 17 13:20:56 2000 Delivered-To: freebsd-stable@freebsd.org Received: from trinity.skynet.be (trinity.skynet.be [195.238.2.38]) by hub.freebsd.org (Postfix) with ESMTP id BE26637B844 for ; Thu, 17 Feb 2000 13:20:48 -0800 (PST) (envelope-from blk@skynet.be) Received: from [195.238.1.121] (brad.techos.skynet.be [195.238.1.121]) by trinity.skynet.be (Postfix) with ESMTP id B220B122F8; Thu, 17 Feb 2000 22:20:42 +0100 (MET) Mime-Version: 1.0 X-Sender: blk@pop.skynet.be Message-Id: In-Reply-To: <20000217103516.C19043@lava.net> References: <20000217103516.C19043@lava.net> Date: Thu, 17 Feb 2000 22:03:47 +0100 To: Clifton Royston From: Brad Knowles Subject: Re: Initial performance testing w/ postmark & softupdates... Cc: Tom , freebsd-stable@FreeBSD.ORG Content-Type: text/plain; charset="us-ascii" ; format="flowed" Sender: owner-freebsd-stable@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG At 10:35 AM -1000 2000/2/17, Clifton Royston wrote: > I guess the real question as to whether that is a fair comparison, is > whether softupdates is getting to the level of predictability and > absolutely 100% recoverability for unexpected shutdowns or crashes > which is expected of the current crop of NFS servers. I disagree that you get 100% guarantees of 100% recoverability for unexpected shutdowns or crashes. Maybe I'm wrong, but I've just seen too many weird problems with trying to write NFS data and not having it work correctly. > That's the kind of performance that a > Netapp or an EMC Celerra is currently promising. Given my past experience with NetApp and EMC, I don't buy it. However, if I were to buy it, I'd want those kinds of features on "local" drives (perhaps connected to a SAN), and I wouldn't want to waste time and effort doing them through NFS. > I'll take a look. Joe always has interesting things to say; but I > don't know that a news spool server necessarily has the same design > priorities as an NFS server. Again, not saying the idea is unworkable, > just urging a little caution. Of course, he's doing RAID-0 across the board for speed on his filesystems for news articles, and no thought given to reliability because he's got multiple servers for that kind of redundancy. However, while software RAID-5 may or may not be rock-solid under vinum, I believe that you could implement it with high-end Mylex, AMI, or DPT controllers and so long as the arrays are configured along the lines that Joe lays out and the controllers have enough battery-backed write-back cache, then you ought to be able to make RAID-5 work well enough for you. Me, I'm going to try a drive array from a competitor of EMC, and see if we can get the reliability, speed, etc... that I'd like to see. We've had some bad experiences with them so far, but then we configured that array ourselves, as opposed to having their experts come in and do it. For now, I'm willing to give them the benefit of the doubt. I'll post updates to my web pages once the benchmarks are done on the array. -- These are my opinions and should not be taken as official Skynet policy _________________________________________________________________________ |o| Brad Knowles, Belgacom Skynet NV/SA |o| |o| Systems Architect, Mail/News/FTP/Proxy Admin Rue Col. Bourg, 124 |o| |o| Phone/Fax: +32-2-706.13.11/726.93.11 B-1140 Brussels |o| |o| http://www.skynet.be Belgium |o| \/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/ Unix is like a wigwam -- no Gates, no Windows, and an Apache inside. Unix is very user-friendly. It's just picky who its friends are. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-stable" in the body of the message