Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!linus!philabs!cmcl2!seismo!gatech!gitpyr!kludge
From: kludge@gitpyr.UUCP (Scott Dorsey)
Newsgroups: net.arch
Subject: Re: Question on Transputers and virtual memory
Message-ID: <1785@gitpyr.UUCP>
Date: Thu, 15-May-86 22:22:28 EDT
Article-I.D.: gitpyr.1785
Posted: Thu May 15 22:22:28 1986
Date-Received: Sun, 18-May-86 14:14:15 EDT
References: <793@wcwvax.UUCP>
Reply-To: kludge@gitpyr.UUCP (Scott Dorsey)
Organization: Georgia College Of Universal Knowledge
Lines: 32

In article <793@wcwvax.UUCP> rick@wcwvax.UUCP (Richard Filipkiewicz) writes:
>This is just a request for information. Does anybody know of
>Transputer system running in a virtual memory environment? The
>reason I ask is that from the hardware spec. there does not seem
>to be an explicit abort or bus error pin. Perhaps you could use
>the EVENT pin if it stops the currently executing process fast
>enough. A more general version of this question is how to run parallel
>machines in virtual memory.

   Because the Transputer does not allow easy implementation of virtual
memory on the PE, it must be performed by taking advantage of the existance
of multiple PE's.  I have seen a system whereby there were three classes
of memory:  memory local to the processor, mass storage (accessed via a
mass-storage server PE), and common memory.  Common memory is stored on
one PE which has a large amount of memory and a high-speed disk.  If an
area of common memory is needed, it is swapped in, and the block of memory
(segment??) that has lain unused for the longest time (and doesn't have
the sticky bit set) is swapped out.
   The system actually resembles caching more than true virtual memory,
but unusual systems require unusual methods.

-- 
-------
Disclaimer: Everything I say is probably a trademark of someone.  But
            don't worry, I probably don't know what I'm talking about.

Scott Dorsey       " If value corrupts
kaptain_kludge         then absolute value corrupts absolutely"

ICS Programming Lab (Where old terminals go to die), Rich 110,
Georgia Institute of Technology, Box 36681, Atlanta, Georgia 30332
...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!kludge