From mboxrd@z Thu Jan 1 00:00:00 1970 From: lm@mcvoy.com (Larry McVoy) Date: Tue, 21 Feb 2017 19:18:36 -0800 Subject: [TUHS] Mach for i386 / Mt Xinu or other In-Reply-To: References: <20170221120218.E07BA18C10B@mercury.lcs.mit.edu> <20170221164728.GZ20341@mcvoy.com> Message-ID: <20170222031836.GH9439@mcvoy.com> That was pretty early, Ron. I betcha if you tried it now (or 10 years ago) things would be different. On Wed, Feb 22, 2017 at 01:04:16AM +0000, ron minnich wrote: > I got to thinking about the file system sync vs. order argument as a result > of this interesting discussion. > > Back in NJ in 1998 I had a 144-node PC cluster. It started with 80 FreeBSD > nodes, and a few months later we add 64 Linux nodes. It was DARPA funded > and we were looking at issues around clustering. > > What we discovered, when we added the Linux cluster, was that building > power was really terrible. We had not realized it with the freebsd cluster > because it tended to cleanly recover from unplanned nasty power cycles. But > the Linux cluster tended to always have a small number of nodes that did > not get through fsck. > > yeah, it's not that good a data point maybe, but we found it interesting. -- --- Larry McVoy lm at mcvoy.com http://www.mcvoy.com/lm