Newsgroups: comp.lang.dylan,comp.lang.lisp
Path: cantaloupe.srv.cs.cmu.edu!rochester!udel!news.sprintlink.net!noc.netcom.net!netcom.com!NewsWatcher!user
From: hbaker@netcom.com (Henry Baker)
Subject: Re: GC in numerous applications
Message-ID: <hbaker-2410950831330001@10.0.2.15>
Sender: hbaker@netcom7.netcom.com
Organization: nil organization
References: <9509208142.AA814231604@mail.creo.bc.ca> <DERWAY.95Oct23103440@alumni.ndc.com>
Date: Tue, 24 Oct 1995 16:31:33 GMT
Lines: 29
Xref: glinda.oz.cs.cmu.edu comp.lang.dylan:5561 comp.lang.lisp:19685

In article <DERWAY.95Oct23103440@alumni.ndc.com>, derway@ndc.com (D.
Erway) wrote:

> >>>>> "bd" == BDall  <BDall@creo.bc.ca> writes:
> 
>  bd>      There have been numerous postings about the speed of GC versus
>  bd>      'manual' memory management. One target audience which I feel has
>  bd>      been missed is the real-time programmer ...
> 
> Right.  Real time is crucial to about 1/4 of the programming we do.  And it is
> the most difficult 1/4.  It is the area where having a good language like lisp
> or dylan would pay off the most, if we could get acceptable deterministic
> behavior from a GC.
> 
> I think there are results on bounded time GC algorithms.  Anyone?

You might check out the stuff in my ftp/www directory, and also the
stuff at ftp.cs.utexas.edu in /pub/garbage.

> Also, this stuff in the Java pages about GC in a low priority thread sounds
> interesting.  Anyone know anything about it?  It doesn't sound very reasonable
> to me, but what do I know?

The latest IWMM'95 (International Workshop on Memory Management, 1995,
Springer-Verlag (c) 1995) has several papers on real-time gc.

-- 
www/ftp directory:
ftp://ftp.netcom.com/pub/hb/hbaker/home.html
