Path: utzoo!attcan!uunet!lll-winken!lll-tis!ames!pasteur!ucbvax!decwrl!labrea!carcoar!andy
From: andy@carcoar.Stanford.EDU (Andy Freeman)
Newsgroups: comp.arch
Subject: Changing times
Message-ID: <22467@labrea.Stanford.EDU>
Date: 10 Jun 88 17:47:27 GMT
References: <19730@beta.UUCP> <4332@killer.UUCP> <802@elxsi.UUCP> <10078@ames.arc.nasa.gov> <20128@beta.UUCP>
Sender: news@labrea.Stanford.EDU
Reply-To: andy@carcoar.Stanford.EDU (Andy Freeman)
Organization: Stanford University
Lines: 21

In article <20128@beta.UUCP> jlg@beta.UUCP (Jim Giles) writes:
>And (in spite of suggestions) there is NO commercial machine presently
>available which even takes 'hints' about the data usage patterns.

Not long ago, the Los Alamos and Livermore people would write
operating systems; now they won't even modify them.

>								Even
>if such a machine DID exist, the best it could do is tie the speed of
>the non-VM code which does it's own asynchronous paging.

In the best of all possible worlds, this is true.  The typical Los
Alamos/Livermore programmer is not as good as their stars, so I'd bet
that the typical programmer could write more efficient programs on a
hint-using VM system than they currently write using program-specific
I/O.  In the former case, the typical program gets more benefit from
the star's programming talents.

-andy
UUCP:  {arpa gateways, decwrl, uunet, rutgers}!polya.stanford.edu!andy
ARPA:  andy@polya.stanford.edu
(415) 329-1718/723-3088 home/cubicle