[warning: the following may be entirely blue sky , but it is very *nice*
blue sky <g>]

"Concurrent Compacting Garbage Collection of a Persistent Heap"
by James O'Toole, Scott Nettles, and David Gifford. December 1993.
and
"Concurrent Replicating Garbage Collection."
by James O'Toole and Scott Nettles. April 1993.

 http://www.psrg.lcs.mit.edu/publications.html

present a possible approach (real time GC is covered by other papers at this
site).
I very much want to try to convert Squeak over to the replicating GC
approach
presented here (real life has proven extremely time consuming lately
however, sigh).
Replication GC should be capable of at least soft real time constraints
since it has
no (incremental) time dependence on object size or number of objects and its
incremental time slices can be tuned as needed.

 -- Dwight

-----Original Message-----
From: James McCartney <james at clyde.as.utexas.edu>
To: squeak at cs.uiuc.edu <squeak at cs.uiuc.edu>
Date: Friday, April 24, 1998 9:50 PM
Subject: Re: Re[2]: Squeak and SMP


[--snip--]

>
>Do you have a line on a real time concurrent GC that is really
>real time? (Some algorithms have been claimed to be real time
>which aren't.)
>
>
>   --- james mccartney     james at clyde.as.utexas.edu   james at lcsaudio.com