Date: Mon, 30 Nov 1998 16:40:53 -0500 From: "Christopher R. Bowman" <crb@ChrisBowman.com> To: "Jason C. Wells" <jcwells@u.washington.edu> Cc: Mike Smith <mike@smith.net.au>, FreeBSD-chat <freebsd-chat@FreeBSD.ORG> Subject: Re: Diskless Workstations Message-ID: <199811302212.RAA04765@quark.ChrisBowman.com> In-Reply-To: <Pine.BSF.4.05.9811301155550.419-100000@s8-37-26.student.wa shington.edu> References: <199811290843.AAA00385@dingo.cdrom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
At 12:06 PM 11/30/98 -0800, Jason C. Wells wrote: >On Sun, 29 Nov 1998, Mike Smith wrote: > >>> Jason Wrote: >>> Sometime ago I asked folks about NC's. Now I am back. Once again I am >>> armed with just enough info to be dangerous. >> >>Flee! > >:) > >>> It seems that running diskless has a serious advantage of using the same >>> disk space for all of the programs that all the users need. 100baseT can >>> compete with UW-SCSI bit for bit on bandwidth. >> >>No it can't. And NFS doesn't compete for latency. But many users >>don't need that sort of filesystem throughput. > >If 100 Mbps => 80 Mbps then 100bT is as good or better than UW-SCSI on >bandwidth. This is what I based my statement on. It appears that I have a >concept error somehow. The numbers look right to me. Can someone steer me >straight? Uh, where does 80Mbps come from? SCSI is 5 Mhz, Fast SCSI is 10Mhz, Ultra is 20 Mhz, times 1 byte for Narrow or 2 bytes for Wide. So Ultra and Wide SCSI should be 40 MByes/sec or 320 Mbps not 80Mbps. Maybe you are thinking U2W which is 80 MBps or 640Mbps. So 100BTX isn't better than UltraWide SCSI. Of course this all ignores transaction overhead, packet overhead, inter packet space and realistic models of the latencies and bandwidths of the devices on the other end. It is just a rough theoretical max throughput. -------- Christopher R. Bowman crb@ChrisBowman.com http://www.ChrisBowman.com/ To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199811302212.RAA04765>