> On a slightly related note, I think I've heard it discussed somewhere at
> some point about chaining VAXstations together over their SCSI busses. It
> may have been with one of the VAX/Linux people.  In the case of the
> VS3100's with dual SCSI controllers would it be possible to say take two
> machines and patch their SCSI/B (2nd) SCSI controllers together?  Then set
> up each controller with a unique SCSI ID on the shared bus.  Then you
> could build a driver for the SCSI devices to make them look like
> point-to-point network interfaces.  Basically use them for passing data
> between VAXen at "high speed".  On top of this you could use something
> like PVM or MPI to run parallel computations across the machines.  And
> then lets say you take the dual SCSI controller machines, you chain then
> together in a big long bus topology, or even a big ring :-) 
> 
> So, is this a technically feasible abuse of SCSI?  And if so, how hard
> would it be to make driver for the SCSI devices that made them look like
> network interfaces?  AND if we could intersperse a SCSI disk drive or two
> in the chain, we might have a faster alternative to ethernet for shared
> disk storage.

Hi Brian,

You might want to read RFC2143 for inspiration. The introduction is
included below.

Brian.

Network Working Group                                        B. Elliston
Request for Comments: 2143                             Compucat Research
Category: Experimental                                          May 1997

       Encapsulating IP with the Small Computer System Interface

1.  Introduction

   As the capacity of local area networks increases to meet the demands
   of high volume application data, there is a class of network
   intensive problems which may be applied to small clusters of
   workstations with high bandwidth interconnection.

   A general observation of networks is that the bit rate of the data
   path typically decreases as the distance between hosts increases.  It
   is common to see regional networks connected at a rate of 64Kbps and
   office networks connected at 100Mbps, but the inverse is far less
   common.

   The same is true of peripheral and memory interconnection.  Memory
   close to a CPU's core may run at speeds equivalent to a gigabit
   network.  More importantly, devices such as disks may connect a
   number of metres away from its host at speeds well in excess of
   current local area network capacity.

   This document outlines a protocol for connecting hosts running the
   TCP/IP protocol suite over a Small Computer System Interface (SCSI)
   bus.  Despite the limitation in the furthest distance between hosts,
   SCSI permits close clusters of workstations to communicate between
   each other at speeds approaching 360 megabits per second.

--
-----------------------------------
Brian Murray      Proximity Pty Ltd
http://www.proximity.com.au/~brian/
-----------------------------------