Re: Help in DSM design

From: Richard B. Johnson (root@chaos.analogic.com)
Date: Sat Mar 04 2000 - 19:54:40 EST


On Fri, 3 Mar 2000, Ronald G. Minnich wrote:

> On Fri, 3 Mar 2000, Richard B. Johnson wrote:
>
> > On Sat, 4 Mar 2000, Richard Gooch wrote:
> >
> > >
> > > Ach! Not another DSM project :-( Don't do it. There's already a DSM
> >
> > [dsm horror story]
>
> I swore I would avoid this argument but ...
>
> look, there's DSMs and there's DSMs. And there's different types of
> applications. I've done a fair number of parallel apps and there are cases
> where you can use DSM, and cases where you can't and you use message
> passing. I can point to failures with both models.
>
> You want to see failures with DSM, you'll find 'em, because people do
> stupid things with them. That doesn't automatically condemn the
> technology, it just means it's not universal (what is?).
>
> You want to see a few successes, well, look at my web page and check the
> papers.
>

Don't get all bent out of shape. The point is that if you don't need
DSM (i.e., Customer requirement), it's not a very good idea to use it.

There are many "papers", mostly written by those who "finally got
something to work...", that will show the greatness of practically
anything. Reference: "Stockholm Syndrome".

Cheers,
Dick Johnson

Penguin : Linux version 2.3.41 on an i686 machine (800.63 BogoMips).

-
To unsubscribe from this list: send the line "unsubscribe linux-net" in
the body of a message to majordomo@vger.rutgers.edu



This archive was generated by hypermail 2b29 : Tue Mar 07 2000 - 21:00:27 EST