Path: utzoo!attcan!uunet!husc6!bloom-beacon!mit-eddie!ll-xn!ames!lll-tis!lll-winken!lll-crg.llnl.gov!brooks
From: brooks@lll-crg.llnl.gov (Eugene D. Brooks III)
Newsgroups: comp.arch
Subject: Re: Is Shared Memory Necessary?
Message-ID: <7331@lll-winken.llnl.gov>
Date: 16 May 88 21:44:45 GMT
References: <503@xios.XIOS.UUCP> <2676@pdn.UUCP> <674@cernvax.UUCP> <9559@sol.ARPA> <685@thalia.rice.edu>
Sender: usenet@lll-winken.llnl.gov
Reply-To: brooks@lll-crg.llnl.gov.UUCP (Eugene D. Brooks III)
Organization: Lawrence Livermore National Laboratory
Lines: 12

In article <685@thalia.rice.edu> retrac@rice.edu (John Carter) writes:
>I think that it's currently worth the cost, but I doubt seriously that it'll
>be a realistic approach as we try to increase the number of processors beyond
>several dozen.  A shared (hardware) memory that needed to handle thousands of
Experience with the 30 CPU Sequent Symmetry system indicates that the
limit for bus based systems with the right copy/back cache protocol
is beyond 30 processors.

Experience with the Cerberus multiprocessor simulator indicates that
a non-bus shared memory system can be very successfully used with
processor counts of several hundred and beyond without fancy fetch-and-op
support in the memory subsystem.  Give me a couple hundred Cray class
processors and I would be happy for at least a week. :-)