Date: Fri, 10 May 1996 00:38:00 -0700 (PDT) From: asami@cs.berkeley.edu (Satoshi Asami) To: msmith@atrad.adelaide.edu.au Cc: terry@lambert.org, msmith@atrad.adelaide.edu.au, current@FreeBSD.org Subject: Re: NFS in -current is _BUSTED_ Message-ID: <199605100738.AAA24652@silvia.HIP.Berkeley.EDU> In-Reply-To: <199605100625.PAA07996@genesis.atrad.adelaide.edu.au> (message from Michael Smith on Fri, 10 May 1996 15:55:17 %2B0930 (CST))
next in thread | previous in thread | raw e-mail | index | archive | help
* > What, someone's actually using THAT in their kernel??? Or libc??? * * Different bcopy optimisations. Diff -stable i386/i386/support.s against * the -current version to see what Terry's talking about. Really? The bcopy code hasn't changed for a long time, it's December that it last changed. I thought NFS was stable until pretty recently.... * > Satoshi "please don't blame me for everything that's going wrong" Asami * * Au contraire. So how big _is_ this filesystem? I was being told by a * rampant Linux-fanatic genetecist the other day that "of the PC unices, * only Linux could possibly manage either of the HGI database, because * it's so big." * * According to him it's around the 100GB mark; obviously you wouldn't So how does Linux build a filesystem that size? Do they have a striped disk array driver too? * put this on one filesystem for performance/backup reasons, but it would * be very funny to counter his drivel... 8) Well, we hit one limit at 128GB yesterday (32 x 4GB). I modified <sys/disklabel.h> to support disk numbers greater than 32 (more on this later), and right now it's at 160GB (40 x 4GB). We don't have any more disks right now. But fear not, we're getting 40 more so I'll post here when we get to 320GB. :) Satoshi
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199605100738.AAA24652>