Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Path: utzoo!mnetor!uunet!seismo!ll-xn!ames!amdcad!amdahl!chuck From: chuck@amdahl.amdahl.com (Charles Simmons) Newsgroups: comp.arch Subject: Re: how fast could disks be (Mach 1) Message-ID: <12357@amdahl.amdahl.com> Date: Mon, 17-Aug-87 13:01:05 EDT Article-I.D.: amdahl.12357 Posted: Mon Aug 17 13:01:05 1987 Date-Received: Tue, 18-Aug-87 05:29:39 EDT References: <12191@amdahl.amdahl.com> <1189@k.gp.cs.cmu.edu> <1075@vlsi.cs.cmu.edu> Reply-To: chuck@amdahl.amdahl.com (Charles Simmons) Organization: Amdahl Corp, Sunnyvale CA Lines: 31 Keywords: laser disk scan Mach In article <1075@vlsi.cs.cmu.edu> gwu@vlsi.cs.cmu.edu (George Wu) writes: > > There's been all this talk about gaining faster mass storage with >parallel operations, and more recently, optical disks. But it seems to me that >a limiting factor are the mechanical parts, which must physically move. Even >using a scanning laser beam to read optical disks, a mirror must be >manipulated, unless someone has found a way to bend light. Relativity anybody? > > But what about non-mechanical storage systems? Are there feasible >magnetic and/or solid state systems out there? I remember there used to be a >big hoopla about bubble memory, but I haven't heard anything recently. And I'm >not sure why, but intuitively, I feel that solid state devices would be even >faster than magnetic ones. Is this just a bias left over from magnetic tapes >and drives? > > George How about an optical disk with a laser driven by a phased array (as in phased array radar)? (Is there such a thing as a phased array laser?) Or maybe some other storage medium accessed by a phased array? Such a creature would have no moving parts. A few months ago, Byte magazine had an article on a British company that was working on wafer scale integration. There first product was supposed to be a 7 Megabyte solid-state "disk drive". Access to this memory would be serial, but seek and transfer times were on the order of 1000 times faster than seek and transfer times on vanilla disks. -- Chuck